a8cd0c01a6
Model: fpadovani/heb-hebr-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.269087026161238,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.68021821975708,
|
|
"epoch": 0.005339028296849973,
|
|
"grad_norm": 31.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 13.9711,
|
|
"mean_token_accuracy": 0.009074668399989605,
|
|
"num_tokens": 10658.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.704092264175415,
|
|
"epoch": 0.010678056593699947,
|
|
"grad_norm": 32.75,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 13.9298,
|
|
"mean_token_accuracy": 0.00866024480201304,
|
|
"num_tokens": 21327.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.7672216415405275,
|
|
"epoch": 0.01601708489054992,
|
|
"grad_norm": 36.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.7823,
|
|
"mean_token_accuracy": 0.007704685814678669,
|
|
"num_tokens": 32403.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.958004379272461,
|
|
"epoch": 0.021356113187399894,
|
|
"grad_norm": 40.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3459,
|
|
"mean_token_accuracy": 0.006649625021964312,
|
|
"num_tokens": 43471.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.477138376235962,
|
|
"epoch": 0.026695141484249868,
|
|
"grad_norm": 38.75,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.4632,
|
|
"mean_token_accuracy": 0.0020860590098891406,
|
|
"num_tokens": 53724.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.8245813846588135,
|
|
"epoch": 0.03203416978109984,
|
|
"grad_norm": 30.375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.6429,
|
|
"mean_token_accuracy": 0.0007934643130283803,
|
|
"num_tokens": 65115.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.22093849182129,
|
|
"epoch": 0.03737319807794982,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.8521,
|
|
"mean_token_accuracy": 0.0002344618726056069,
|
|
"num_tokens": 75902.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.684853076934814,
|
|
"epoch": 0.04271222637479979,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.7564,
|
|
"mean_token_accuracy": 0.0011873395298607648,
|
|
"num_tokens": 87051.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.69392204284668,
|
|
"epoch": 0.04805125467164976,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.5885,
|
|
"mean_token_accuracy": 0.006381610780954361,
|
|
"num_tokens": 98998.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.600316715240478,
|
|
"epoch": 0.053390282968499736,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.4445,
|
|
"mean_token_accuracy": 0.011505381856113672,
|
|
"num_tokens": 111207.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.673304462432862,
|
|
"epoch": 0.05872931126534971,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2638,
|
|
"mean_token_accuracy": 0.016619019117206334,
|
|
"num_tokens": 121039.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.670036125183106,
|
|
"epoch": 0.06406833956219968,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1533,
|
|
"mean_token_accuracy": 0.015064884908497334,
|
|
"num_tokens": 132604.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.671291160583497,
|
|
"epoch": 0.06940736785904965,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0181,
|
|
"mean_token_accuracy": 0.03150726184248924,
|
|
"num_tokens": 142825.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.688156223297119,
|
|
"epoch": 0.07474639615589963,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9346,
|
|
"mean_token_accuracy": 0.056240572780370715,
|
|
"num_tokens": 153121.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.670577144622802,
|
|
"epoch": 0.0800854244527496,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.8809,
|
|
"mean_token_accuracy": 0.05582823716104031,
|
|
"num_tokens": 164675.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.664248943328857,
|
|
"epoch": 0.08542445274959957,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8352,
|
|
"mean_token_accuracy": 0.04962811656296253,
|
|
"num_tokens": 175146.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.648977661132813,
|
|
"epoch": 0.09076348104644955,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8124,
|
|
"mean_token_accuracy": 0.04548483826220036,
|
|
"num_tokens": 186177.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.642164325714111,
|
|
"epoch": 0.09610250934329952,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.685,
|
|
"mean_token_accuracy": 0.048678891733288765,
|
|
"num_tokens": 197143.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.636518096923828,
|
|
"epoch": 0.10144153764014949,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.6182,
|
|
"mean_token_accuracy": 0.052110004052519796,
|
|
"num_tokens": 207979.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.610146522521973,
|
|
"epoch": 0.10678056593699947,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.5135,
|
|
"mean_token_accuracy": 0.0568073820322752,
|
|
"num_tokens": 218733.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.569573020935058,
|
|
"epoch": 0.11211959423384944,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.4144,
|
|
"mean_token_accuracy": 0.05693732798099518,
|
|
"num_tokens": 228887.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.517056083679199,
|
|
"epoch": 0.11745862253069941,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.3275,
|
|
"mean_token_accuracy": 0.05374570451676845,
|
|
"num_tokens": 240289.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.480204105377197,
|
|
"epoch": 0.12279765082754938,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.2538,
|
|
"mean_token_accuracy": 0.05631314069032669,
|
|
"num_tokens": 250548.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.422281837463379,
|
|
"epoch": 0.12813667912439936,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.1616,
|
|
"mean_token_accuracy": 0.05645494796335697,
|
|
"num_tokens": 261669.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.368197917938232,
|
|
"epoch": 0.13347570742124934,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.0945,
|
|
"mean_token_accuracy": 0.05810702405869961,
|
|
"num_tokens": 272926.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.291021633148194,
|
|
"epoch": 0.1388147357180993,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.0261,
|
|
"mean_token_accuracy": 0.05110162906348705,
|
|
"num_tokens": 284173.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.235691547393799,
|
|
"epoch": 0.14415376401494928,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.9153,
|
|
"mean_token_accuracy": 0.05686397515237331,
|
|
"num_tokens": 294850.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.190592193603516,
|
|
"epoch": 0.14949279231179927,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.8582,
|
|
"mean_token_accuracy": 0.05346960201859474,
|
|
"num_tokens": 305727.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.10048542022705,
|
|
"epoch": 0.15483182060864922,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.7462,
|
|
"mean_token_accuracy": 0.058417992666363716,
|
|
"num_tokens": 317243.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.970914077758788,
|
|
"epoch": 0.1601708489054992,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.7387,
|
|
"mean_token_accuracy": 0.05338601246476173,
|
|
"num_tokens": 328334.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.846445274353027,
|
|
"epoch": 0.16550987720234916,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.6482,
|
|
"mean_token_accuracy": 0.054720787703990935,
|
|
"num_tokens": 339774.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.73973569869995,
|
|
"epoch": 0.17084890549919915,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5543,
|
|
"mean_token_accuracy": 0.05573372282087803,
|
|
"num_tokens": 350211.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.550614070892333,
|
|
"epoch": 0.1761879337960491,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4684,
|
|
"mean_token_accuracy": 0.054642071947455406,
|
|
"num_tokens": 360109.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.380690956115723,
|
|
"epoch": 0.1815269620928991,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4131,
|
|
"mean_token_accuracy": 0.054827965795993805,
|
|
"num_tokens": 370911.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.282039070129395,
|
|
"epoch": 0.18686599038974908,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3991,
|
|
"mean_token_accuracy": 0.060196897760033605,
|
|
"num_tokens": 381808.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.13407211303711,
|
|
"epoch": 0.19220501868659903,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3091,
|
|
"mean_token_accuracy": 0.06106638610363006,
|
|
"num_tokens": 392175.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.000042152404784,
|
|
"epoch": 0.19754404698344902,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3874,
|
|
"mean_token_accuracy": 0.049684269726276396,
|
|
"num_tokens": 403261.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.931573295593262,
|
|
"epoch": 0.20288307528029897,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.2853,
|
|
"mean_token_accuracy": 0.05462104417383671,
|
|
"num_tokens": 414992.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.791546726226807,
|
|
"epoch": 0.20822210357714896,
|
|
"grad_norm": 0.6328125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2732,
|
|
"mean_token_accuracy": 0.05595530979335308,
|
|
"num_tokens": 425316.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.729821968078614,
|
|
"epoch": 0.21356113187399894,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.344,
|
|
"mean_token_accuracy": 0.05437092892825603,
|
|
"num_tokens": 437759.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.712938117980958,
|
|
"epoch": 0.2189001601708489,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.2844,
|
|
"mean_token_accuracy": 0.05534115061163902,
|
|
"num_tokens": 448066.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.651909160614014,
|
|
"epoch": 0.22423918846769889,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.2206,
|
|
"mean_token_accuracy": 0.055615832656621934,
|
|
"num_tokens": 458632.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.63439655303955,
|
|
"epoch": 0.22957821676454884,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.2606,
|
|
"mean_token_accuracy": 0.054173830151557925,
|
|
"num_tokens": 469224.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.596088218688966,
|
|
"epoch": 0.23491724506139883,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2823,
|
|
"mean_token_accuracy": 0.05661116987466812,
|
|
"num_tokens": 480492.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.625334930419921,
|
|
"epoch": 0.2402562733582488,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.2398,
|
|
"mean_token_accuracy": 0.057639367505908014,
|
|
"num_tokens": 491325.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.569988918304443,
|
|
"epoch": 0.24559530165509877,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2465,
|
|
"mean_token_accuracy": 0.0603976845741272,
|
|
"num_tokens": 502240.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.592293834686279,
|
|
"epoch": 0.25093432995194875,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2035,
|
|
"mean_token_accuracy": 0.06263627484440804,
|
|
"num_tokens": 512710.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.58193883895874,
|
|
"epoch": 0.2562733582487987,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.2513,
|
|
"mean_token_accuracy": 0.05789706408977509,
|
|
"num_tokens": 524023.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.547989177703858,
|
|
"epoch": 0.26161238654564867,
|
|
"grad_norm": 3.234375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.1422,
|
|
"mean_token_accuracy": 0.06462252773344516,
|
|
"num_tokens": 534612.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.475037765502929,
|
|
"epoch": 0.2669514148424987,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2036,
|
|
"mean_token_accuracy": 0.05881951674818993,
|
|
"num_tokens": 544866.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.527704906463622,
|
|
"epoch": 0.27229044313934864,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2523,
|
|
"mean_token_accuracy": 0.055548783764243125,
|
|
"num_tokens": 556091.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.524747467041015,
|
|
"epoch": 0.2776294714361986,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2536,
|
|
"mean_token_accuracy": 0.05479630120098591,
|
|
"num_tokens": 566946.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.54754581451416,
|
|
"epoch": 0.2829684997330486,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.1371,
|
|
"mean_token_accuracy": 0.060486416518688205,
|
|
"num_tokens": 577164.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.525337314605713,
|
|
"epoch": 0.28830752802989856,
|
|
"grad_norm": 0.671875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.1768,
|
|
"mean_token_accuracy": 0.05711319632828236,
|
|
"num_tokens": 588744.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.505339241027832,
|
|
"epoch": 0.2936465563267485,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.1727,
|
|
"mean_token_accuracy": 0.055232659354805945,
|
|
"num_tokens": 600943.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.492932891845703,
|
|
"epoch": 0.29898558462359853,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.1726,
|
|
"mean_token_accuracy": 0.06307518370449543,
|
|
"num_tokens": 611864.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.451044940948487,
|
|
"epoch": 0.3043246129204485,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.152,
|
|
"mean_token_accuracy": 0.05931071788072586,
|
|
"num_tokens": 623314.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.428306388854981,
|
|
"epoch": 0.30966364121729845,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.1763,
|
|
"mean_token_accuracy": 0.06025952883064747,
|
|
"num_tokens": 633509.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.451824760437011,
|
|
"epoch": 0.3150026695141484,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.0983,
|
|
"mean_token_accuracy": 0.06582859642803669,
|
|
"num_tokens": 644655.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.374590492248535,
|
|
"epoch": 0.3203416978109984,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.135,
|
|
"mean_token_accuracy": 0.05644083172082901,
|
|
"num_tokens": 655940.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.408124351501465,
|
|
"epoch": 0.3256807261078484,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.1809,
|
|
"mean_token_accuracy": 0.056555739417672155,
|
|
"num_tokens": 667074.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.498903751373291,
|
|
"epoch": 0.33101975440469833,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2309,
|
|
"mean_token_accuracy": 0.06025502309203148,
|
|
"num_tokens": 677942.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.455632591247559,
|
|
"epoch": 0.33635878270154834,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.1642,
|
|
"mean_token_accuracy": 0.06168182455003261,
|
|
"num_tokens": 689226.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.403513431549072,
|
|
"epoch": 0.3416978109983983,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.1786,
|
|
"mean_token_accuracy": 0.059801208600401876,
|
|
"num_tokens": 700830.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.456379795074463,
|
|
"epoch": 0.34703683929524826,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1377,
|
|
"mean_token_accuracy": 0.06001129299402237,
|
|
"num_tokens": 712677.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.363616943359375,
|
|
"epoch": 0.3523758675920982,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2395,
|
|
"mean_token_accuracy": 0.05579938478767872,
|
|
"num_tokens": 723833.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.497436428070069,
|
|
"epoch": 0.3577148958889482,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1771,
|
|
"mean_token_accuracy": 0.059333324804902075,
|
|
"num_tokens": 735526.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.438710403442382,
|
|
"epoch": 0.3630539241857982,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1223,
|
|
"mean_token_accuracy": 0.06136825978755951,
|
|
"num_tokens": 745934.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.391829586029052,
|
|
"epoch": 0.36839295248264814,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.0837,
|
|
"mean_token_accuracy": 0.06251248680055141,
|
|
"num_tokens": 756693.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.393558979034424,
|
|
"epoch": 0.37373198077949815,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.1418,
|
|
"mean_token_accuracy": 0.05746800377964974,
|
|
"num_tokens": 767236.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.491841030120849,
|
|
"epoch": 0.3790710090763481,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1254,
|
|
"mean_token_accuracy": 0.06065751984715462,
|
|
"num_tokens": 778552.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.428674793243408,
|
|
"epoch": 0.38441003737319807,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1788,
|
|
"mean_token_accuracy": 0.05917464569211006,
|
|
"num_tokens": 789615.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.48459882736206,
|
|
"epoch": 0.3897490656700481,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.1355,
|
|
"mean_token_accuracy": 0.06367055512964725,
|
|
"num_tokens": 799998.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.322556686401366,
|
|
"epoch": 0.39508809396689804,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.0679,
|
|
"mean_token_accuracy": 0.062298642471432686,
|
|
"num_tokens": 810042.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.422898483276366,
|
|
"epoch": 0.400427122263748,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0755,
|
|
"mean_token_accuracy": 0.05906784273684025,
|
|
"num_tokens": 821692.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.36357364654541,
|
|
"epoch": 0.40576615056059795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1164,
|
|
"mean_token_accuracy": 0.06748287416994572,
|
|
"num_tokens": 832262.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.333745193481445,
|
|
"epoch": 0.41110517885744796,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.04,
|
|
"mean_token_accuracy": 0.06151870898902416,
|
|
"num_tokens": 843006.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.353145503997803,
|
|
"epoch": 0.4164442071542979,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0332,
|
|
"mean_token_accuracy": 0.062286530062556265,
|
|
"num_tokens": 853978.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.357436656951904,
|
|
"epoch": 0.4217832354511479,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0743,
|
|
"mean_token_accuracy": 0.05977092795073986,
|
|
"num_tokens": 864634.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.423241233825683,
|
|
"epoch": 0.4271222637479979,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.0568,
|
|
"mean_token_accuracy": 0.06662314645946026,
|
|
"num_tokens": 875379.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.398690319061279,
|
|
"epoch": 0.43246129204484784,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.0674,
|
|
"mean_token_accuracy": 0.06020735241472721,
|
|
"num_tokens": 886428.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.425256729125977,
|
|
"epoch": 0.4378003203416978,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0827,
|
|
"mean_token_accuracy": 0.0645828690379858,
|
|
"num_tokens": 896700.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.369815444946289,
|
|
"epoch": 0.44313934863854776,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0577,
|
|
"mean_token_accuracy": 0.06499057859182358,
|
|
"num_tokens": 907478.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.39303207397461,
|
|
"epoch": 0.44847837693539777,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9707,
|
|
"mean_token_accuracy": 0.07010989226400852,
|
|
"num_tokens": 918170.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.28305778503418,
|
|
"epoch": 0.45381740523224773,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.0345,
|
|
"mean_token_accuracy": 0.06685064136981964,
|
|
"num_tokens": 928866.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.279892349243164,
|
|
"epoch": 0.4591564335290977,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0205,
|
|
"mean_token_accuracy": 0.06419647410511971,
|
|
"num_tokens": 940395.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.475063610076905,
|
|
"epoch": 0.4644954618259477,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0365,
|
|
"mean_token_accuracy": 0.062472108751535416,
|
|
"num_tokens": 950814.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.192297077178955,
|
|
"epoch": 0.46983449012279765,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.0225,
|
|
"mean_token_accuracy": 0.0661535993218422,
|
|
"num_tokens": 961658.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.43046703338623,
|
|
"epoch": 0.4751735184196476,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0993,
|
|
"mean_token_accuracy": 0.06110986284911633,
|
|
"num_tokens": 974790.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.331879615783691,
|
|
"epoch": 0.4805125467164976,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.9772,
|
|
"mean_token_accuracy": 0.06222324036061764,
|
|
"num_tokens": 986599.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.243362045288086,
|
|
"epoch": 0.4858515750133476,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.9787,
|
|
"mean_token_accuracy": 0.06861450895667076,
|
|
"num_tokens": 997731.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.290034008026122,
|
|
"epoch": 0.49119060331019754,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0374,
|
|
"mean_token_accuracy": 0.06714494079351425,
|
|
"num_tokens": 1009343.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.351651859283447,
|
|
"epoch": 0.4965296316070475,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.9541,
|
|
"mean_token_accuracy": 0.0648241501301527,
|
|
"num_tokens": 1019878.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.248888206481933,
|
|
"epoch": 0.5018686599038975,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.9553,
|
|
"mean_token_accuracy": 0.07038948647677898,
|
|
"num_tokens": 1030551.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.30935525894165,
|
|
"epoch": 0.5072076882007475,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9893,
|
|
"mean_token_accuracy": 0.06879530549049377,
|
|
"num_tokens": 1041432.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.283312702178955,
|
|
"epoch": 0.5125467164975974,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9704,
|
|
"mean_token_accuracy": 0.06568947955965995,
|
|
"num_tokens": 1051769.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.256877326965332,
|
|
"epoch": 0.5178857447944474,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9282,
|
|
"mean_token_accuracy": 0.06400737687945365,
|
|
"num_tokens": 1062334.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.27625093460083,
|
|
"epoch": 0.5232247730912973,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.921,
|
|
"mean_token_accuracy": 0.06769514791667461,
|
|
"num_tokens": 1072792.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.173498821258544,
|
|
"epoch": 0.5285638013881474,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.8148,
|
|
"mean_token_accuracy": 0.07047795243561268,
|
|
"num_tokens": 1083642.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.192062282562256,
|
|
"epoch": 0.5339028296849974,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9111,
|
|
"mean_token_accuracy": 0.07375151067972183,
|
|
"num_tokens": 1094454.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5339028296849974,
|
|
"eval_entropy": 8.031556909734553,
|
|
"eval_loss": 8.007415771484375,
|
|
"eval_mean_token_accuracy": 0.06165650523715205,
|
|
"eval_num_tokens": 1094454.0,
|
|
"eval_runtime": 1.8278,
|
|
"eval_samples_per_second": 1685.073,
|
|
"eval_steps_per_second": 210.634,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.316486740112305,
|
|
"epoch": 0.5392418579818473,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9525,
|
|
"mean_token_accuracy": 0.06484691724181176,
|
|
"num_tokens": 1105068.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.111512756347656,
|
|
"epoch": 0.5445808862786973,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9097,
|
|
"mean_token_accuracy": 0.07020260021090508,
|
|
"num_tokens": 1115668.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.238842296600343,
|
|
"epoch": 0.5499199145755472,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8867,
|
|
"mean_token_accuracy": 0.07027992233633995,
|
|
"num_tokens": 1126742.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.179580593109131,
|
|
"epoch": 0.5552589428723972,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9668,
|
|
"mean_token_accuracy": 0.0645244125276804,
|
|
"num_tokens": 1137351.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.22506046295166,
|
|
"epoch": 0.5605979711692471,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.8967,
|
|
"mean_token_accuracy": 0.0711884018033743,
|
|
"num_tokens": 1148389.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.167451763153077,
|
|
"epoch": 0.5659369994660972,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.8994,
|
|
"mean_token_accuracy": 0.0703209575265646,
|
|
"num_tokens": 1158798.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.239407634735107,
|
|
"epoch": 0.5712760277629472,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.948,
|
|
"mean_token_accuracy": 0.06852544359862804,
|
|
"num_tokens": 1169005.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.212612915039063,
|
|
"epoch": 0.5766150560597971,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.8255,
|
|
"mean_token_accuracy": 0.06702273786067962,
|
|
"num_tokens": 1180560.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.105984878540038,
|
|
"epoch": 0.5819540843566471,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8035,
|
|
"mean_token_accuracy": 0.07094659134745598,
|
|
"num_tokens": 1191162.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.162831687927246,
|
|
"epoch": 0.587293112653497,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8252,
|
|
"mean_token_accuracy": 0.06873422190546989,
|
|
"num_tokens": 1202017.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.132958889007568,
|
|
"epoch": 0.592632140950347,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8232,
|
|
"mean_token_accuracy": 0.07796628661453724,
|
|
"num_tokens": 1213356.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.137973403930664,
|
|
"epoch": 0.5979711692471971,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8335,
|
|
"mean_token_accuracy": 0.07087706178426742,
|
|
"num_tokens": 1224164.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.08298101425171,
|
|
"epoch": 0.603310197544047,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8463,
|
|
"mean_token_accuracy": 0.0711651612073183,
|
|
"num_tokens": 1235483.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.127214622497558,
|
|
"epoch": 0.608649225840897,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8595,
|
|
"mean_token_accuracy": 0.07112185694277287,
|
|
"num_tokens": 1248013.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.093348598480224,
|
|
"epoch": 0.6139882541377469,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.7999,
|
|
"mean_token_accuracy": 0.07371774725615979,
|
|
"num_tokens": 1258512.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.157558727264405,
|
|
"epoch": 0.6193272824345969,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8256,
|
|
"mean_token_accuracy": 0.06662675254046917,
|
|
"num_tokens": 1269425.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.97422046661377,
|
|
"epoch": 0.6246663107314468,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.7958,
|
|
"mean_token_accuracy": 0.07273220643401146,
|
|
"num_tokens": 1280595.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.247371101379395,
|
|
"epoch": 0.6300053390282968,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8485,
|
|
"mean_token_accuracy": 0.06550373584032058,
|
|
"num_tokens": 1290873.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.015791416168213,
|
|
"epoch": 0.6353443673251469,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.852,
|
|
"mean_token_accuracy": 0.066574127972126,
|
|
"num_tokens": 1303150.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.053708696365357,
|
|
"epoch": 0.6406833956219968,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.768,
|
|
"mean_token_accuracy": 0.07012191005051135,
|
|
"num_tokens": 1314862.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.02850046157837,
|
|
"epoch": 0.6460224239188468,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8029,
|
|
"mean_token_accuracy": 0.07391900308430195,
|
|
"num_tokens": 1326052.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.128786611557008,
|
|
"epoch": 0.6513614522156967,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.7667,
|
|
"mean_token_accuracy": 0.0753765020519495,
|
|
"num_tokens": 1336847.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.02261905670166,
|
|
"epoch": 0.6567004805125467,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.7701,
|
|
"mean_token_accuracy": 0.06919390633702278,
|
|
"num_tokens": 1347611.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.006579780578614,
|
|
"epoch": 0.6620395088093967,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7859,
|
|
"mean_token_accuracy": 0.07407167293131352,
|
|
"num_tokens": 1358679.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.048807954788208,
|
|
"epoch": 0.6673785371062466,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7849,
|
|
"mean_token_accuracy": 0.07473289892077446,
|
|
"num_tokens": 1369409.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.044457149505615,
|
|
"epoch": 0.6727175654030967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7618,
|
|
"mean_token_accuracy": 0.06960076540708542,
|
|
"num_tokens": 1380380.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.039245176315308,
|
|
"epoch": 0.6780565936999466,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7081,
|
|
"mean_token_accuracy": 0.07184483371675014,
|
|
"num_tokens": 1390885.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.9913702487945555,
|
|
"epoch": 0.6833956219967966,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7418,
|
|
"mean_token_accuracy": 0.07358726263046264,
|
|
"num_tokens": 1400960.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.986459016799927,
|
|
"epoch": 0.6887346502936466,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.6917,
|
|
"mean_token_accuracy": 0.07739836499094963,
|
|
"num_tokens": 1411484.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.129097318649292,
|
|
"epoch": 0.6940736785904965,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7908,
|
|
"mean_token_accuracy": 0.07160034552216529,
|
|
"num_tokens": 1422931.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.91273045539856,
|
|
"epoch": 0.6994127068873465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7237,
|
|
"mean_token_accuracy": 0.07407537326216698,
|
|
"num_tokens": 1434347.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.050477170944214,
|
|
"epoch": 0.7047517351841964,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.8135,
|
|
"mean_token_accuracy": 0.07437525913119317,
|
|
"num_tokens": 1445687.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.082316970825195,
|
|
"epoch": 0.7100907634810465,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.6471,
|
|
"mean_token_accuracy": 0.08003439009189606,
|
|
"num_tokens": 1456335.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.9566230297088625,
|
|
"epoch": 0.7154297917778965,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7071,
|
|
"mean_token_accuracy": 0.07267946116626263,
|
|
"num_tokens": 1467369.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.986024713516235,
|
|
"epoch": 0.7207688200747464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7612,
|
|
"mean_token_accuracy": 0.06941658817231655,
|
|
"num_tokens": 1478734.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.984552240371704,
|
|
"epoch": 0.7261078483715964,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7359,
|
|
"mean_token_accuracy": 0.07321545220911503,
|
|
"num_tokens": 1489807.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.967883205413818,
|
|
"epoch": 0.7314468766684463,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6794,
|
|
"mean_token_accuracy": 0.07043348215520381,
|
|
"num_tokens": 1501646.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.948320388793945,
|
|
"epoch": 0.7367859049652963,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.6987,
|
|
"mean_token_accuracy": 0.07335374914109707,
|
|
"num_tokens": 1511849.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.999287462234497,
|
|
"epoch": 0.7421249332621462,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7004,
|
|
"mean_token_accuracy": 0.07678697407245635,
|
|
"num_tokens": 1522183.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.895766305923462,
|
|
"epoch": 0.7474639615589963,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7063,
|
|
"mean_token_accuracy": 0.07557001523673534,
|
|
"num_tokens": 1532819.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.02945156097412,
|
|
"epoch": 0.7528029898558463,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6718,
|
|
"mean_token_accuracy": 0.07557161971926689,
|
|
"num_tokens": 1543909.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.9088582515716555,
|
|
"epoch": 0.7581420181526962,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7087,
|
|
"mean_token_accuracy": 0.07322404161095619,
|
|
"num_tokens": 1554147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.878606176376342,
|
|
"epoch": 0.7634810464495462,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6136,
|
|
"mean_token_accuracy": 0.07956673577427864,
|
|
"num_tokens": 1565466.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.8776044845581055,
|
|
"epoch": 0.7688200747463961,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.07377611920237541,
|
|
"num_tokens": 1577264.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.885418844223023,
|
|
"epoch": 0.7741591030432461,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.686,
|
|
"mean_token_accuracy": 0.0757551722228527,
|
|
"num_tokens": 1588222.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.901127099990845,
|
|
"epoch": 0.7794981313400962,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6224,
|
|
"mean_token_accuracy": 0.07221109680831432,
|
|
"num_tokens": 1600609.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.929693412780762,
|
|
"epoch": 0.7848371596369461,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.669,
|
|
"mean_token_accuracy": 0.07948594614863395,
|
|
"num_tokens": 1612032.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.876828384399414,
|
|
"epoch": 0.7901761879337961,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6973,
|
|
"mean_token_accuracy": 0.0742558129131794,
|
|
"num_tokens": 1623228.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.857063436508179,
|
|
"epoch": 0.795515216230646,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6469,
|
|
"mean_token_accuracy": 0.07817403376102447,
|
|
"num_tokens": 1634293.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.910915946960449,
|
|
"epoch": 0.800854244527496,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6183,
|
|
"mean_token_accuracy": 0.08015101104974746,
|
|
"num_tokens": 1645782.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.885407543182373,
|
|
"epoch": 0.8061932728243459,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.5871,
|
|
"mean_token_accuracy": 0.0775064155459404,
|
|
"num_tokens": 1657060.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.78002610206604,
|
|
"epoch": 0.8115323011211959,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.568,
|
|
"mean_token_accuracy": 0.08284709006547927,
|
|
"num_tokens": 1668017.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.908759117126465,
|
|
"epoch": 0.816871329418046,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6398,
|
|
"mean_token_accuracy": 0.07324047535657882,
|
|
"num_tokens": 1678331.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.884358358383179,
|
|
"epoch": 0.8222103577148959,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.5591,
|
|
"mean_token_accuracy": 0.0813971921801567,
|
|
"num_tokens": 1688719.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.755726432800293,
|
|
"epoch": 0.8275493860117459,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.5821,
|
|
"mean_token_accuracy": 0.07562258839607239,
|
|
"num_tokens": 1699028.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.950335073471069,
|
|
"epoch": 0.8328884143085958,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5798,
|
|
"mean_token_accuracy": 0.07690282315015792,
|
|
"num_tokens": 1711122.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.790361976623535,
|
|
"epoch": 0.8382274426054458,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.5939,
|
|
"mean_token_accuracy": 0.07726500257849693,
|
|
"num_tokens": 1721694.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.733095169067383,
|
|
"epoch": 0.8435664709022958,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5939,
|
|
"mean_token_accuracy": 0.0780683733522892,
|
|
"num_tokens": 1732530.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.877895975112915,
|
|
"epoch": 0.8489054991991457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5577,
|
|
"mean_token_accuracy": 0.08093490153551101,
|
|
"num_tokens": 1742860.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.750740194320679,
|
|
"epoch": 0.8542445274959958,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5315,
|
|
"mean_token_accuracy": 0.0812942422926426,
|
|
"num_tokens": 1752900.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.801487588882447,
|
|
"epoch": 0.8595835557928457,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5465,
|
|
"mean_token_accuracy": 0.07882912531495094,
|
|
"num_tokens": 1763293.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.752074146270752,
|
|
"epoch": 0.8649225840896957,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5696,
|
|
"mean_token_accuracy": 0.0819572426378727,
|
|
"num_tokens": 1773893.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.815144634246826,
|
|
"epoch": 0.8702616123865456,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.6358,
|
|
"mean_token_accuracy": 0.07636146545410157,
|
|
"num_tokens": 1785568.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.8184651851654055,
|
|
"epoch": 0.8756006406833956,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5594,
|
|
"mean_token_accuracy": 0.08072886392474174,
|
|
"num_tokens": 1797387.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.813911390304566,
|
|
"epoch": 0.8809396689802456,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5902,
|
|
"mean_token_accuracy": 0.07879998087882996,
|
|
"num_tokens": 1808382.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.829354333877563,
|
|
"epoch": 0.8862786972770955,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5745,
|
|
"mean_token_accuracy": 0.0824002705514431,
|
|
"num_tokens": 1818937.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.799869441986084,
|
|
"epoch": 0.8916177255739456,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.5099,
|
|
"mean_token_accuracy": 0.07985455431044101,
|
|
"num_tokens": 1830547.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.7153465270996096,
|
|
"epoch": 0.8969567538707955,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.5052,
|
|
"mean_token_accuracy": 0.08278301656246186,
|
|
"num_tokens": 1841569.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.764185190200806,
|
|
"epoch": 0.9022957821676455,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5154,
|
|
"mean_token_accuracy": 0.08583573549985886,
|
|
"num_tokens": 1852300.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.718573999404907,
|
|
"epoch": 0.9076348104644955,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5077,
|
|
"mean_token_accuracy": 0.08417970538139344,
|
|
"num_tokens": 1863664.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.6863548278808596,
|
|
"epoch": 0.9129738387613454,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.4426,
|
|
"mean_token_accuracy": 0.08229955136775971,
|
|
"num_tokens": 1873623.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.687994432449341,
|
|
"epoch": 0.9183128670581954,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5429,
|
|
"mean_token_accuracy": 0.0791929729282856,
|
|
"num_tokens": 1886092.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.751056241989136,
|
|
"epoch": 0.9236518953550453,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5525,
|
|
"mean_token_accuracy": 0.08293391354382038,
|
|
"num_tokens": 1897983.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.769723224639892,
|
|
"epoch": 0.9289909236518954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5134,
|
|
"mean_token_accuracy": 0.08427821174263954,
|
|
"num_tokens": 1908193.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.7698462963104244,
|
|
"epoch": 0.9343299519487454,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.4921,
|
|
"mean_token_accuracy": 0.08160046115517616,
|
|
"num_tokens": 1918628.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.668918085098267,
|
|
"epoch": 0.9396689802455953,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5372,
|
|
"mean_token_accuracy": 0.08014441058039665,
|
|
"num_tokens": 1929071.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.654262733459473,
|
|
"epoch": 0.9450080085424453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5003,
|
|
"mean_token_accuracy": 0.08210862651467324,
|
|
"num_tokens": 1939567.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.773276090621948,
|
|
"epoch": 0.9503470368392952,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5402,
|
|
"mean_token_accuracy": 0.08018058091402054,
|
|
"num_tokens": 1951165.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.798144149780273,
|
|
"epoch": 0.9556860651361452,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.4594,
|
|
"mean_token_accuracy": 0.08885749988257885,
|
|
"num_tokens": 1960855.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.707589864730835,
|
|
"epoch": 0.9610250934329952,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.4564,
|
|
"mean_token_accuracy": 0.08240535110235214,
|
|
"num_tokens": 1971873.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.671180057525635,
|
|
"epoch": 0.9663641217298452,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4848,
|
|
"mean_token_accuracy": 0.08260536044836045,
|
|
"num_tokens": 1983561.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.731744003295899,
|
|
"epoch": 0.9717031500266952,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4828,
|
|
"mean_token_accuracy": 0.08393343463540077,
|
|
"num_tokens": 1994534.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.590392780303955,
|
|
"epoch": 0.9770421783235451,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.432,
|
|
"mean_token_accuracy": 0.08598880358040333,
|
|
"num_tokens": 2005121.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.696979951858521,
|
|
"epoch": 0.9823812066203951,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4211,
|
|
"mean_token_accuracy": 0.08999745771288872,
|
|
"num_tokens": 2016260.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.615106248855591,
|
|
"epoch": 0.987720234917245,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4074,
|
|
"mean_token_accuracy": 0.08892037123441696,
|
|
"num_tokens": 2025705.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.7097069263458256,
|
|
"epoch": 0.993059263214095,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5596,
|
|
"mean_token_accuracy": 0.07989178895950318,
|
|
"num_tokens": 2036675.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.544618844985962,
|
|
"epoch": 0.9983982915109451,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4114,
|
|
"mean_token_accuracy": 0.08617193698883056,
|
|
"num_tokens": 2047877.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.8020609219868975,
|
|
"epoch": 1.0032034169781099,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.3332,
|
|
"mean_token_accuracy": 0.08596346527338028,
|
|
"num_tokens": 2057153.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.623947048187256,
|
|
"epoch": 1.00854244527496,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4568,
|
|
"mean_token_accuracy": 0.0771791972219944,
|
|
"num_tokens": 2068760.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.589405059814453,
|
|
"epoch": 1.01388147357181,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.3104,
|
|
"mean_token_accuracy": 0.0916103646159172,
|
|
"num_tokens": 2078671.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.588209199905395,
|
|
"epoch": 1.01922050186866,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.3495,
|
|
"mean_token_accuracy": 0.08704460337758065,
|
|
"num_tokens": 2089581.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.6823986053466795,
|
|
"epoch": 1.02455953016551,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.293,
|
|
"mean_token_accuracy": 0.09483219981193543,
|
|
"num_tokens": 2100026.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.51586127281189,
|
|
"epoch": 1.02989855846236,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.3062,
|
|
"mean_token_accuracy": 0.09631321728229522,
|
|
"num_tokens": 2111838.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.603687572479248,
|
|
"epoch": 1.0352375867592098,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.3908,
|
|
"mean_token_accuracy": 0.0855435535311699,
|
|
"num_tokens": 2123063.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.658422660827637,
|
|
"epoch": 1.0405766150560598,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.279,
|
|
"mean_token_accuracy": 0.09217825904488564,
|
|
"num_tokens": 2134321.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.596205329895019,
|
|
"epoch": 1.0459156433529098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.3405,
|
|
"mean_token_accuracy": 0.09226287379860879,
|
|
"num_tokens": 2144216.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.586561298370361,
|
|
"epoch": 1.0512546716497597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3236,
|
|
"mean_token_accuracy": 0.08801514208316803,
|
|
"num_tokens": 2155757.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.509102058410645,
|
|
"epoch": 1.0565936999466097,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2762,
|
|
"mean_token_accuracy": 0.08867826834321021,
|
|
"num_tokens": 2166661.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.554547595977783,
|
|
"epoch": 1.0619327282434596,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.2522,
|
|
"mean_token_accuracy": 0.0905068688094616,
|
|
"num_tokens": 2176796.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.565032482147217,
|
|
"epoch": 1.0672717565403096,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.3151,
|
|
"mean_token_accuracy": 0.09214983209967613,
|
|
"num_tokens": 2186821.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0672717565403096,
|
|
"eval_entropy": 7.661522679514699,
|
|
"eval_loss": 7.484097957611084,
|
|
"eval_mean_token_accuracy": 0.08444042449163926,
|
|
"eval_num_tokens": 2186821.0,
|
|
"eval_runtime": 1.8185,
|
|
"eval_samples_per_second": 1693.67,
|
|
"eval_steps_per_second": 211.709,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.488584995269775,
|
|
"epoch": 1.0726107848371595,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999997458093253,
|
|
"loss": 7.2872,
|
|
"mean_token_accuracy": 0.08735908269882202,
|
|
"num_tokens": 2197953.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.587278175354004,
|
|
"epoch": 1.0779498131340097,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999987131606938,
|
|
"loss": 7.2946,
|
|
"mean_token_accuracy": 0.09318367019295692,
|
|
"num_tokens": 2208033.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.508203601837158,
|
|
"epoch": 1.0832888414308597,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999968861708311,
|
|
"loss": 7.306,
|
|
"mean_token_accuracy": 0.0833747498691082,
|
|
"num_tokens": 2218010.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.607468748092652,
|
|
"epoch": 1.0886278697277096,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999942648461873,
|
|
"loss": 7.3213,
|
|
"mean_token_accuracy": 0.08709867373108864,
|
|
"num_tokens": 2229079.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.554487895965576,
|
|
"epoch": 1.0939668980245596,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999908491960169,
|
|
"loss": 7.3188,
|
|
"mean_token_accuracy": 0.08554268255829811,
|
|
"num_tokens": 2240333.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.41667366027832,
|
|
"epoch": 1.0993059263214096,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999866392323784,
|
|
"loss": 7.2619,
|
|
"mean_token_accuracy": 0.09279655292630196,
|
|
"num_tokens": 2251385.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.586219072341919,
|
|
"epoch": 1.1046449546182595,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999816349701349,
|
|
"loss": 7.2562,
|
|
"mean_token_accuracy": 0.09476209431886673,
|
|
"num_tokens": 2261890.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.429596185684204,
|
|
"epoch": 1.1099839829151095,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999758364269535,
|
|
"loss": 7.287,
|
|
"mean_token_accuracy": 0.0902016930282116,
|
|
"num_tokens": 2272510.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.436788702011109,
|
|
"epoch": 1.1153230112119594,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999692436233055,
|
|
"loss": 7.2791,
|
|
"mean_token_accuracy": 0.08760200515389442,
|
|
"num_tokens": 2283392.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.570240783691406,
|
|
"epoch": 1.1206620395088094,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999618565824665,
|
|
"loss": 7.2279,
|
|
"mean_token_accuracy": 0.09378096982836723,
|
|
"num_tokens": 2295155.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.490195608139038,
|
|
"epoch": 1.1260010678056593,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999536753305157,
|
|
"loss": 7.3268,
|
|
"mean_token_accuracy": 0.08983222171664237,
|
|
"num_tokens": 2307208.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.629471921920777,
|
|
"epoch": 1.1313400961025093,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999446998963364,
|
|
"loss": 7.2481,
|
|
"mean_token_accuracy": 0.09013928584754467,
|
|
"num_tokens": 2317480.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.449492406845093,
|
|
"epoch": 1.1366791243993593,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999349303116159,
|
|
"loss": 7.3373,
|
|
"mean_token_accuracy": 0.08976641297340393,
|
|
"num_tokens": 2328840.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.580901098251343,
|
|
"epoch": 1.1420181526962092,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999243666108448,
|
|
"loss": 7.285,
|
|
"mean_token_accuracy": 0.09045098423957824,
|
|
"num_tokens": 2340450.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.4406897068023685,
|
|
"epoch": 1.1473571809930592,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999130088313176,
|
|
"loss": 7.3033,
|
|
"mean_token_accuracy": 0.09590284526348114,
|
|
"num_tokens": 2351415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.485285329818725,
|
|
"epoch": 1.1526962092899091,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999008570131319,
|
|
"loss": 7.2965,
|
|
"mean_token_accuracy": 0.09630362093448638,
|
|
"num_tokens": 2362153.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.625531721115112,
|
|
"epoch": 1.1580352375867593,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998879111991889,
|
|
"loss": 7.2759,
|
|
"mean_token_accuracy": 0.089786946028471,
|
|
"num_tokens": 2372352.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.374940299987793,
|
|
"epoch": 1.1633742658836093,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499874171435193,
|
|
"loss": 7.2659,
|
|
"mean_token_accuracy": 0.09284015148878097,
|
|
"num_tokens": 2382294.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.572332715988159,
|
|
"epoch": 1.1687132941804592,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998596377696511,
|
|
"loss": 7.2382,
|
|
"mean_token_accuracy": 0.09612747579813004,
|
|
"num_tokens": 2394054.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.367642450332641,
|
|
"epoch": 1.1740523224773092,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998443102538734,
|
|
"loss": 7.2585,
|
|
"mean_token_accuracy": 0.09028631299734116,
|
|
"num_tokens": 2405408.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.572077369689941,
|
|
"epoch": 1.1793913507741591,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998281889419725,
|
|
"loss": 7.2905,
|
|
"mean_token_accuracy": 0.08611830845475196,
|
|
"num_tokens": 2416378.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.381543397903442,
|
|
"epoch": 1.184730379071009,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998112738908636,
|
|
"loss": 7.183,
|
|
"mean_token_accuracy": 0.09558979794383049,
|
|
"num_tokens": 2427016.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.501509714126587,
|
|
"epoch": 1.190069407367859,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997935651602641,
|
|
"loss": 7.1302,
|
|
"mean_token_accuracy": 0.09728729203343392,
|
|
"num_tokens": 2437824.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.4092631340026855,
|
|
"epoch": 1.195408435664709,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997750628126931,
|
|
"loss": 7.2588,
|
|
"mean_token_accuracy": 0.08902693465352059,
|
|
"num_tokens": 2448829.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.444784927368164,
|
|
"epoch": 1.200747463961559,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997557669134718,
|
|
"loss": 7.2186,
|
|
"mean_token_accuracy": 0.09314063340425491,
|
|
"num_tokens": 2459629.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.362234878540039,
|
|
"epoch": 1.206086492258409,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004997356775307231,
|
|
"loss": 7.2803,
|
|
"mean_token_accuracy": 0.09419681504368782,
|
|
"num_tokens": 2471478.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.532480144500733,
|
|
"epoch": 1.2114255205552589,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499714794735371,
|
|
"loss": 7.1266,
|
|
"mean_token_accuracy": 0.10543641149997711,
|
|
"num_tokens": 2481573.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.350391530990601,
|
|
"epoch": 1.2167645488521088,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996931186011406,
|
|
"loss": 7.256,
|
|
"mean_token_accuracy": 0.09152917265892029,
|
|
"num_tokens": 2493429.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.421112728118897,
|
|
"epoch": 1.222103577148959,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499670649204558,
|
|
"loss": 7.218,
|
|
"mean_token_accuracy": 0.09748484790325165,
|
|
"num_tokens": 2503882.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.419456720352173,
|
|
"epoch": 1.227442605445809,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996473866249496,
|
|
"loss": 7.1971,
|
|
"mean_token_accuracy": 0.09453025832772255,
|
|
"num_tokens": 2513978.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.512584495544433,
|
|
"epoch": 1.232781633742659,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996233309444424,
|
|
"loss": 7.2582,
|
|
"mean_token_accuracy": 0.09497607797384262,
|
|
"num_tokens": 2523806.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.397188282012939,
|
|
"epoch": 1.2381206620395089,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499598482247963,
|
|
"loss": 7.2416,
|
|
"mean_token_accuracy": 0.09581049084663391,
|
|
"num_tokens": 2533695.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.449572372436523,
|
|
"epoch": 1.2434596903363588,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995728406232378,
|
|
"loss": 7.1616,
|
|
"mean_token_accuracy": 0.10089872553944587,
|
|
"num_tokens": 2544315.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.433754110336304,
|
|
"epoch": 1.2487987186332088,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499546406160793,
|
|
"loss": 7.1648,
|
|
"mean_token_accuracy": 0.09890817850828171,
|
|
"num_tokens": 2555985.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.336302137374878,
|
|
"epoch": 1.2541377469300587,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004995191789539531,
|
|
"loss": 7.2444,
|
|
"mean_token_accuracy": 0.09039352908730507,
|
|
"num_tokens": 2567472.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.473226404190063,
|
|
"epoch": 1.2594767752269087,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004994911590988421,
|
|
"loss": 7.2288,
|
|
"mean_token_accuracy": 0.09728409722447395,
|
|
"num_tokens": 2578030.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.461479711532593,
|
|
"epoch": 1.2648158035237587,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994623466943816,
|
|
"loss": 7.2471,
|
|
"mean_token_accuracy": 0.093011786788702,
|
|
"num_tokens": 2589925.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.40461311340332,
|
|
"epoch": 1.2701548318206086,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994327418422922,
|
|
"loss": 7.1787,
|
|
"mean_token_accuracy": 0.09393652454018593,
|
|
"num_tokens": 2601825.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.403125238418579,
|
|
"epoch": 1.2754938601174586,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994023446470912,
|
|
"loss": 7.1985,
|
|
"mean_token_accuracy": 0.09799938648939133,
|
|
"num_tokens": 2612726.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.348568964004516,
|
|
"epoch": 1.2808328884143085,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004993711552160936,
|
|
"loss": 7.1336,
|
|
"mean_token_accuracy": 0.09304829761385917,
|
|
"num_tokens": 2623753.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.404272222518921,
|
|
"epoch": 1.2861719167111585,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993391736594117,
|
|
"loss": 7.1593,
|
|
"mean_token_accuracy": 0.0968549020588398,
|
|
"num_tokens": 2635275.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.480840063095092,
|
|
"epoch": 1.2915109450080084,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993064000899538,
|
|
"loss": 7.2019,
|
|
"mean_token_accuracy": 0.09882493540644646,
|
|
"num_tokens": 2646329.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.275875473022461,
|
|
"epoch": 1.2968499733048584,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004992728346234246,
|
|
"loss": 7.2084,
|
|
"mean_token_accuracy": 0.09495408535003662,
|
|
"num_tokens": 2657629.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.453315734863281,
|
|
"epoch": 1.3021890016017084,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992384773783244,
|
|
"loss": 7.1886,
|
|
"mean_token_accuracy": 0.09616189375519753,
|
|
"num_tokens": 2667746.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.27887864112854,
|
|
"epoch": 1.3075280298985585,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004992033284759491,
|
|
"loss": 7.1507,
|
|
"mean_token_accuracy": 0.10439181700348854,
|
|
"num_tokens": 2678157.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.484576034545898,
|
|
"epoch": 1.3128670581954085,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991673880403891,
|
|
"loss": 7.2097,
|
|
"mean_token_accuracy": 0.09509778320789337,
|
|
"num_tokens": 2689319.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.382059717178345,
|
|
"epoch": 1.3182060864922585,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004991306561985294,
|
|
"loss": 7.1406,
|
|
"mean_token_accuracy": 0.09976356402039528,
|
|
"num_tokens": 2699718.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.359973287582397,
|
|
"epoch": 1.3235451147891084,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004990931330800492,
|
|
"loss": 7.19,
|
|
"mean_token_accuracy": 0.09778250828385353,
|
|
"num_tokens": 2710888.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.3708795547485355,
|
|
"epoch": 1.3288841430859584,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990548188174211,
|
|
"loss": 7.1108,
|
|
"mean_token_accuracy": 0.0972976379096508,
|
|
"num_tokens": 2722024.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.322368574142456,
|
|
"epoch": 1.3342231713828083,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990157135459105,
|
|
"loss": 7.0693,
|
|
"mean_token_accuracy": 0.09633990824222564,
|
|
"num_tokens": 2731780.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.395527935028076,
|
|
"epoch": 1.3395621996796583,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004989758174035758,
|
|
"loss": 7.1745,
|
|
"mean_token_accuracy": 0.0960126519203186,
|
|
"num_tokens": 2742524.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.409232425689697,
|
|
"epoch": 1.3449012279765082,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004989351305312676,
|
|
"loss": 7.1378,
|
|
"mean_token_accuracy": 0.09832295700907707,
|
|
"num_tokens": 2752077.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.322344446182251,
|
|
"epoch": 1.3502402562733582,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988936530726276,
|
|
"loss": 7.1473,
|
|
"mean_token_accuracy": 0.09714090898633003,
|
|
"num_tokens": 2763625.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.293940925598145,
|
|
"epoch": 1.3555792845702082,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004988513851740889,
|
|
"loss": 7.1327,
|
|
"mean_token_accuracy": 0.10281604751944542,
|
|
"num_tokens": 2775262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.3934331893920895,
|
|
"epoch": 1.3609183128670583,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004988083269848753,
|
|
"loss": 7.1964,
|
|
"mean_token_accuracy": 0.09679302722215652,
|
|
"num_tokens": 2786449.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.385905408859253,
|
|
"epoch": 1.3662573411639083,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004987644786570003,
|
|
"loss": 7.1696,
|
|
"mean_token_accuracy": 0.09244814142584801,
|
|
"num_tokens": 2796891.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.33904447555542,
|
|
"epoch": 1.3715963694607582,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004987198403452674,
|
|
"loss": 7.1607,
|
|
"mean_token_accuracy": 0.09559275880455971,
|
|
"num_tokens": 2806926.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.3863444328308105,
|
|
"epoch": 1.3769353977576082,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004986744122072688,
|
|
"loss": 7.0899,
|
|
"mean_token_accuracy": 0.09752057045698166,
|
|
"num_tokens": 2817750.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.180061292648316,
|
|
"epoch": 1.3822744260544582,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004986281944033851,
|
|
"loss": 7.1001,
|
|
"mean_token_accuracy": 0.10428606942296029,
|
|
"num_tokens": 2827770.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.401736879348755,
|
|
"epoch": 1.3876134543513081,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004985811870967848,
|
|
"loss": 7.1361,
|
|
"mean_token_accuracy": 0.10217376500368118,
|
|
"num_tokens": 2839583.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.363368988037109,
|
|
"epoch": 1.392952482648158,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 7.1358,
|
|
"mean_token_accuracy": 0.10174108073115348,
|
|
"num_tokens": 2850621.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.260199737548828,
|
|
"epoch": 1.398291510945008,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004984848046420445,
|
|
"loss": 7.1302,
|
|
"mean_token_accuracy": 0.09183874055743217,
|
|
"num_tokens": 2862287.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.392441749572754,
|
|
"epoch": 1.403630539241858,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004984354298341756,
|
|
"loss": 7.1278,
|
|
"mean_token_accuracy": 0.10441029965877532,
|
|
"num_tokens": 2871769.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.305068445205689,
|
|
"epoch": 1.408969567538708,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004983852662041312,
|
|
"loss": 7.1009,
|
|
"mean_token_accuracy": 0.0979885146021843,
|
|
"num_tokens": 2882590.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.323603248596191,
|
|
"epoch": 1.414308595835558,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00049833431392901,
|
|
"loss": 7.1618,
|
|
"mean_token_accuracy": 0.09200560078024864,
|
|
"num_tokens": 2893903.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.257247972488403,
|
|
"epoch": 1.4196476241324079,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982825731886956,
|
|
"loss": 7.0838,
|
|
"mean_token_accuracy": 0.09896045774221421,
|
|
"num_tokens": 2904532.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.321243238449097,
|
|
"epoch": 1.4249866524292578,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004982300441658546,
|
|
"loss": 7.0602,
|
|
"mean_token_accuracy": 0.09538981467485427,
|
|
"num_tokens": 2916340.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.252778673171997,
|
|
"epoch": 1.4303256807261078,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004981767270459367,
|
|
"loss": 7.0686,
|
|
"mean_token_accuracy": 0.10086248964071273,
|
|
"num_tokens": 2926424.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.384427738189697,
|
|
"epoch": 1.4356647090229577,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004981226220171744,
|
|
"loss": 7.1487,
|
|
"mean_token_accuracy": 0.09660428166389465,
|
|
"num_tokens": 2937620.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.306267547607422,
|
|
"epoch": 1.4410037373198077,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004980677292705811,
|
|
"loss": 7.0283,
|
|
"mean_token_accuracy": 0.10887585952877998,
|
|
"num_tokens": 2948274.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.251709699630737,
|
|
"epoch": 1.4463427656166576,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004980120489999519,
|
|
"loss": 7.1382,
|
|
"mean_token_accuracy": 0.09639003127813339,
|
|
"num_tokens": 2958986.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.4081559658050535,
|
|
"epoch": 1.4516817939135078,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004979555814018617,
|
|
"loss": 7.162,
|
|
"mean_token_accuracy": 0.09467695504426957,
|
|
"num_tokens": 2970933.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.283700609207154,
|
|
"epoch": 1.4570208222103578,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000497898326675665,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.10104830637574196,
|
|
"num_tokens": 2981197.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.330152797698974,
|
|
"epoch": 1.4623598505072077,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004978402850234956,
|
|
"loss": 7.1262,
|
|
"mean_token_accuracy": 0.10599048286676407,
|
|
"num_tokens": 2993732.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.264380216598511,
|
|
"epoch": 1.4676988788040577,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004977814566502651,
|
|
"loss": 7.071,
|
|
"mean_token_accuracy": 0.10518286749720573,
|
|
"num_tokens": 3004617.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.271407747268677,
|
|
"epoch": 1.4730379071009077,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004977218417636629,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.10099183842539787,
|
|
"num_tokens": 3015058.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.180662345886231,
|
|
"epoch": 1.4783769353977576,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004976614405741546,
|
|
"loss": 7.0199,
|
|
"mean_token_accuracy": 0.10256859883666039,
|
|
"num_tokens": 3025995.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.355836534500122,
|
|
"epoch": 1.4837159636946076,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004976002532949823,
|
|
"loss": 7.0472,
|
|
"mean_token_accuracy": 0.1040610671043396,
|
|
"num_tokens": 3036060.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.186236763000489,
|
|
"epoch": 1.4890549919914575,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004975382801421631,
|
|
"loss": 7.0578,
|
|
"mean_token_accuracy": 0.1070075087249279,
|
|
"num_tokens": 3047012.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.2215704917907715,
|
|
"epoch": 1.4943940202883075,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004974755213344886,
|
|
"loss": 6.9844,
|
|
"mean_token_accuracy": 0.1064968764781952,
|
|
"num_tokens": 3058186.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.222527360916137,
|
|
"epoch": 1.4997330485851574,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004974119770935242,
|
|
"loss": 6.9946,
|
|
"mean_token_accuracy": 0.10440263673663139,
|
|
"num_tokens": 3068566.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.2857832431793215,
|
|
"epoch": 1.5050720768820076,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004973476476436081,
|
|
"loss": 7.0199,
|
|
"mean_token_accuracy": 0.10458874180912972,
|
|
"num_tokens": 3079588.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.08985447883606,
|
|
"epoch": 1.5104111051788576,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972825332118505,
|
|
"loss": 7.0023,
|
|
"mean_token_accuracy": 0.10629142746329308,
|
|
"num_tokens": 3092376.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.245244026184082,
|
|
"epoch": 1.5157501334757075,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004972166340281331,
|
|
"loss": 7.0461,
|
|
"mean_token_accuracy": 0.10235923826694489,
|
|
"num_tokens": 3103439.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.26457109451294,
|
|
"epoch": 1.5210891617725575,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004971499503251083,
|
|
"loss": 7.1008,
|
|
"mean_token_accuracy": 0.10314854681491852,
|
|
"num_tokens": 3115007.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.262210988998413,
|
|
"epoch": 1.5264281900694074,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004970824823381976,
|
|
"loss": 7.0539,
|
|
"mean_token_accuracy": 0.10062195062637329,
|
|
"num_tokens": 3125632.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.242954254150391,
|
|
"epoch": 1.5317672183662574,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000497014230305592,
|
|
"loss": 7.0316,
|
|
"mean_token_accuracy": 0.10418535098433494,
|
|
"num_tokens": 3135825.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.2584309577941895,
|
|
"epoch": 1.5371062466631074,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004969451944682501,
|
|
"loss": 7.0369,
|
|
"mean_token_accuracy": 0.10161466002464295,
|
|
"num_tokens": 3147689.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.136695671081543,
|
|
"epoch": 1.5424452749599573,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004968753750698978,
|
|
"loss": 6.993,
|
|
"mean_token_accuracy": 0.10334522351622581,
|
|
"num_tokens": 3158754.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.298619794845581,
|
|
"epoch": 1.5477843032568073,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004968047723570273,
|
|
"loss": 7.0495,
|
|
"mean_token_accuracy": 0.09448943957686425,
|
|
"num_tokens": 3169332.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.218168115615844,
|
|
"epoch": 1.5531233315536572,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004967333865788961,
|
|
"loss": 7.124,
|
|
"mean_token_accuracy": 0.10236010178923607,
|
|
"num_tokens": 3180761.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.4003571510314945,
|
|
"epoch": 1.5584623598505072,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004966612179875267,
|
|
"loss": 7.0762,
|
|
"mean_token_accuracy": 0.09902083575725555,
|
|
"num_tokens": 3191635.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.129079866409302,
|
|
"epoch": 1.5638013881473571,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004965882668377048,
|
|
"loss": 7.0424,
|
|
"mean_token_accuracy": 0.10169995054602624,
|
|
"num_tokens": 3202452.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.34759840965271,
|
|
"epoch": 1.569140416444207,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004965145333869789,
|
|
"loss": 6.971,
|
|
"mean_token_accuracy": 0.10686884224414825,
|
|
"num_tokens": 3213298.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.113559675216675,
|
|
"epoch": 1.574479444741057,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004964400178956595,
|
|
"loss": 7.0056,
|
|
"mean_token_accuracy": 0.10754135549068451,
|
|
"num_tokens": 3224314.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.2728907585144045,
|
|
"epoch": 1.579818473037907,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004963647206268181,
|
|
"loss": 7.0051,
|
|
"mean_token_accuracy": 0.1074408732354641,
|
|
"num_tokens": 3235660.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.222253370285034,
|
|
"epoch": 1.585157501334757,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496288641846286,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.09584841802716255,
|
|
"num_tokens": 3246856.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.304793119430542,
|
|
"epoch": 1.590496529631607,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004962117818226538,
|
|
"loss": 7.1116,
|
|
"mean_token_accuracy": 0.10176649838685989,
|
|
"num_tokens": 3258692.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.17184624671936,
|
|
"epoch": 1.5958355579284569,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00049613414082727,
|
|
"loss": 7.0102,
|
|
"mean_token_accuracy": 0.10286101996898651,
|
|
"num_tokens": 3270577.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.200855493545532,
|
|
"epoch": 1.6011745862253068,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004960557191342405,
|
|
"loss": 6.9376,
|
|
"mean_token_accuracy": 0.10783654004335404,
|
|
"num_tokens": 3281262.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6011745862253068,
|
|
"eval_entropy": 7.068539691281009,
|
|
"eval_loss": 7.1312689781188965,
|
|
"eval_mean_token_accuracy": 0.10161492010796225,
|
|
"eval_num_tokens": 3281262.0,
|
|
"eval_runtime": 1.8156,
|
|
"eval_samples_per_second": 1696.444,
|
|
"eval_steps_per_second": 212.055,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.21118049621582,
|
|
"epoch": 1.606513614522157,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000495976517020427,
|
|
"loss": 6.9955,
|
|
"mean_token_accuracy": 0.10954990237951279,
|
|
"num_tokens": 3291552.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.275783395767212,
|
|
"epoch": 1.611852642819007,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004958965347654469,
|
|
"loss": 7.0386,
|
|
"mean_token_accuracy": 0.10242254436016082,
|
|
"num_tokens": 3301977.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.121581172943115,
|
|
"epoch": 1.617191671115857,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004958157726516715,
|
|
"loss": 6.9699,
|
|
"mean_token_accuracy": 0.10966179370880128,
|
|
"num_tokens": 3312783.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.285579776763916,
|
|
"epoch": 1.622530699412707,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004957342309642256,
|
|
"loss": 6.9964,
|
|
"mean_token_accuracy": 0.10571648105978966,
|
|
"num_tokens": 3323895.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.104770946502685,
|
|
"epoch": 1.6278697277095568,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004956519099909857,
|
|
"loss": 6.9295,
|
|
"mean_token_accuracy": 0.10804956257343293,
|
|
"num_tokens": 3334103.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.208183097839355,
|
|
"epoch": 1.6332087560064068,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004955688100225803,
|
|
"loss": 6.9827,
|
|
"mean_token_accuracy": 0.10653971955180168,
|
|
"num_tokens": 3346279.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.092655944824219,
|
|
"epoch": 1.6385477843032568,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004954849313523875,
|
|
"loss": 6.9509,
|
|
"mean_token_accuracy": 0.10072449594736099,
|
|
"num_tokens": 3357582.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.243341445922852,
|
|
"epoch": 1.643886812600107,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004954002742765347,
|
|
"loss": 6.9531,
|
|
"mean_token_accuracy": 0.10738401338458062,
|
|
"num_tokens": 3369297.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.092299699783325,
|
|
"epoch": 1.649225840896957,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004953148390938973,
|
|
"loss": 7.0684,
|
|
"mean_token_accuracy": 0.10321348384022713,
|
|
"num_tokens": 3382430.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.257068061828614,
|
|
"epoch": 1.6545648691938069,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000495228626106098,
|
|
"loss": 7.0312,
|
|
"mean_token_accuracy": 0.11251498833298683,
|
|
"num_tokens": 3393809.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.158068895339966,
|
|
"epoch": 1.6599038974906568,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004951416356175053,
|
|
"loss": 7.0077,
|
|
"mean_token_accuracy": 0.10660256892442703,
|
|
"num_tokens": 3405164.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.227698040008545,
|
|
"epoch": 1.6652429257875068,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004950538679352327,
|
|
"loss": 7.0212,
|
|
"mean_token_accuracy": 0.09924901276826859,
|
|
"num_tokens": 3416091.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.283638668060303,
|
|
"epoch": 1.6705819540843567,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004949653233691373,
|
|
"loss": 7.0712,
|
|
"mean_token_accuracy": 0.10288623869419097,
|
|
"num_tokens": 3427675.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.135433340072632,
|
|
"epoch": 1.6759209823812067,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004948760022318194,
|
|
"loss": 6.9453,
|
|
"mean_token_accuracy": 0.11070238053798676,
|
|
"num_tokens": 3438119.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.199069452285767,
|
|
"epoch": 1.6812600106780566,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004947859048386205,
|
|
"loss": 6.9806,
|
|
"mean_token_accuracy": 0.10209845751523972,
|
|
"num_tokens": 3449151.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.132083177566528,
|
|
"epoch": 1.6865990389749066,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004946950315076225,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.10959059745073318,
|
|
"num_tokens": 3459697.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.093738555908203,
|
|
"epoch": 1.6919380672717566,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004946033825596474,
|
|
"loss": 6.9422,
|
|
"mean_token_accuracy": 0.10823669135570527,
|
|
"num_tokens": 3470409.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.1705444812774655,
|
|
"epoch": 1.6972770955686065,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004945109583182548,
|
|
"loss": 6.9651,
|
|
"mean_token_accuracy": 0.1007260449230671,
|
|
"num_tokens": 3481515.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.167929172515869,
|
|
"epoch": 1.7026161238654565,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004944177591097415,
|
|
"loss": 6.9069,
|
|
"mean_token_accuracy": 0.10608908087015152,
|
|
"num_tokens": 3492683.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.120382785797119,
|
|
"epoch": 1.7079551521623064,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004943237852631404,
|
|
"loss": 6.9469,
|
|
"mean_token_accuracy": 0.10530265048146248,
|
|
"num_tokens": 3503067.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.19160099029541,
|
|
"epoch": 1.7132941804591564,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004942290371102193,
|
|
"loss": 6.9651,
|
|
"mean_token_accuracy": 0.10647035390138626,
|
|
"num_tokens": 3513611.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.181960487365723,
|
|
"epoch": 1.7186332087560063,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004941335149854794,
|
|
"loss": 6.9617,
|
|
"mean_token_accuracy": 0.10861775949597359,
|
|
"num_tokens": 3523802.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.190891218185425,
|
|
"epoch": 1.7239722370528563,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004940372192261546,
|
|
"loss": 6.9297,
|
|
"mean_token_accuracy": 0.10318318530917167,
|
|
"num_tokens": 3534764.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.121490812301635,
|
|
"epoch": 1.7293112653497063,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004939401501722098,
|
|
"loss": 6.9932,
|
|
"mean_token_accuracy": 0.10756855309009553,
|
|
"num_tokens": 3544852.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.093603610992432,
|
|
"epoch": 1.7346502936465562,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004938423081663401,
|
|
"loss": 6.7933,
|
|
"mean_token_accuracy": 0.11004197001457214,
|
|
"num_tokens": 3555470.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.121168518066407,
|
|
"epoch": 1.7399893219434062,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004937436935539695,
|
|
"loss": 6.9935,
|
|
"mean_token_accuracy": 0.10067067667841911,
|
|
"num_tokens": 3567497.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.194992160797119,
|
|
"epoch": 1.7453283502402561,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004936443066832494,
|
|
"loss": 6.9725,
|
|
"mean_token_accuracy": 0.10623072907328605,
|
|
"num_tokens": 3578354.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.0331216812133786,
|
|
"epoch": 1.7506673785371063,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935441479050579,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.11453714296221733,
|
|
"num_tokens": 3589369.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.2113524913787845,
|
|
"epoch": 1.7560064068339563,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004934432175729981,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10899419113993644,
|
|
"num_tokens": 3600396.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.116479682922363,
|
|
"epoch": 1.7613454351308062,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000493341516043397,
|
|
"loss": 6.9588,
|
|
"mean_token_accuracy": 0.11054658442735672,
|
|
"num_tokens": 3610670.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.107414627075196,
|
|
"epoch": 1.7666844634276562,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004932390436753043,
|
|
"loss": 6.9206,
|
|
"mean_token_accuracy": 0.10563154965639114,
|
|
"num_tokens": 3620821.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.1294042587280275,
|
|
"epoch": 1.7720234917245061,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004931358008304911,
|
|
"loss": 6.8989,
|
|
"mean_token_accuracy": 0.11248534172773361,
|
|
"num_tokens": 3631469.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.078049898147583,
|
|
"epoch": 1.777362520021356,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004930317878734486,
|
|
"loss": 6.9114,
|
|
"mean_token_accuracy": 0.1081713691353798,
|
|
"num_tokens": 3641605.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.201143026351929,
|
|
"epoch": 1.782701548318206,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004929270051713866,
|
|
"loss": 6.9417,
|
|
"mean_token_accuracy": 0.1108034186065197,
|
|
"num_tokens": 3652107.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.0569343090057375,
|
|
"epoch": 1.7880405766150562,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004928214530942331,
|
|
"loss": 6.9729,
|
|
"mean_token_accuracy": 0.10586601719260216,
|
|
"num_tokens": 3663399.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.078392410278321,
|
|
"epoch": 1.7933796049119062,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004927151320146313,
|
|
"loss": 6.951,
|
|
"mean_token_accuracy": 0.10743931010365486,
|
|
"num_tokens": 3673638.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.162407350540161,
|
|
"epoch": 1.7987186332087561,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10548929050564766,
|
|
"num_tokens": 3684459.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.04658784866333,
|
|
"epoch": 1.804057661505606,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004925001843522323,
|
|
"loss": 6.9414,
|
|
"mean_token_accuracy": 0.10741329342126846,
|
|
"num_tokens": 3696113.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.169419765472412,
|
|
"epoch": 1.809396689802456,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004923915585282916,
|
|
"loss": 6.9062,
|
|
"mean_token_accuracy": 0.10803867951035499,
|
|
"num_tokens": 3707345.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.098963403701783,
|
|
"epoch": 1.814735718099306,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004922821652196136,
|
|
"loss": 6.744,
|
|
"mean_token_accuracy": 0.11550131812691689,
|
|
"num_tokens": 3717573.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.035619926452637,
|
|
"epoch": 1.820074746396156,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004921720048124034,
|
|
"loss": 6.9383,
|
|
"mean_token_accuracy": 0.11059200093150139,
|
|
"num_tokens": 3726658.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.158073139190674,
|
|
"epoch": 1.825413774693006,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004920610776955742,
|
|
"loss": 6.9495,
|
|
"mean_token_accuracy": 0.11338067129254341,
|
|
"num_tokens": 3737769.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.071769857406617,
|
|
"epoch": 1.8307528029898559,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004919493842607456,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.11217370927333832,
|
|
"num_tokens": 3748608.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.165196466445923,
|
|
"epoch": 1.8360918312867058,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004918369249022433,
|
|
"loss": 6.9148,
|
|
"mean_token_accuracy": 0.11021506115794182,
|
|
"num_tokens": 3759063.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.131874465942383,
|
|
"epoch": 1.8414308595835558,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004917237000170966,
|
|
"loss": 6.8688,
|
|
"mean_token_accuracy": 0.10861963406205177,
|
|
"num_tokens": 3769622.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.098778629302979,
|
|
"epoch": 1.8467698878804057,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004916097100050375,
|
|
"loss": 6.8727,
|
|
"mean_token_accuracy": 0.11213916838169098,
|
|
"num_tokens": 3780587.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.07066969871521,
|
|
"epoch": 1.8521089161772557,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004914949552684992,
|
|
"loss": 6.9236,
|
|
"mean_token_accuracy": 0.11197843477129936,
|
|
"num_tokens": 3792386.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.145204401016235,
|
|
"epoch": 1.8574479444741057,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000491379436212615,
|
|
"loss": 6.9396,
|
|
"mean_token_accuracy": 0.11039775982499123,
|
|
"num_tokens": 3803820.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.06385440826416,
|
|
"epoch": 1.8627869727709556,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000491263153245216,
|
|
"loss": 6.8982,
|
|
"mean_token_accuracy": 0.1090501606464386,
|
|
"num_tokens": 3815187.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.122003555297852,
|
|
"epoch": 1.8681260010678056,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004911461067768309,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10848403200507165,
|
|
"num_tokens": 3826933.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.051048755645752,
|
|
"epoch": 1.8734650293646555,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004910282972206834,
|
|
"loss": 6.7997,
|
|
"mean_token_accuracy": 0.11381994634866714,
|
|
"num_tokens": 3836934.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.05472617149353,
|
|
"epoch": 1.8788040576615055,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004909097249926913,
|
|
"loss": 6.9048,
|
|
"mean_token_accuracy": 0.10961363092064857,
|
|
"num_tokens": 3848662.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.01745228767395,
|
|
"epoch": 1.8841430859583554,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004907903905114652,
|
|
"loss": 6.7385,
|
|
"mean_token_accuracy": 0.1175423063337803,
|
|
"num_tokens": 3860052.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.047714710235596,
|
|
"epoch": 1.8894821142552054,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004906702941983066,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.10923791378736496,
|
|
"num_tokens": 3871867.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.121882438659668,
|
|
"epoch": 1.8948211425520554,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004905494364772065,
|
|
"loss": 6.9453,
|
|
"mean_token_accuracy": 0.1086169384419918,
|
|
"num_tokens": 3883612.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.0241187572479244,
|
|
"epoch": 1.9001601708489055,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004904278177748441,
|
|
"loss": 6.868,
|
|
"mean_token_accuracy": 0.10942395329475403,
|
|
"num_tokens": 3895520.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.096190881729126,
|
|
"epoch": 1.9054991991457555,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004903054385205854,
|
|
"loss": 6.8573,
|
|
"mean_token_accuracy": 0.10636932700872422,
|
|
"num_tokens": 3905898.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.115457010269165,
|
|
"epoch": 1.9108382274426055,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004901822991464811,
|
|
"loss": 6.9076,
|
|
"mean_token_accuracy": 0.10664125084877014,
|
|
"num_tokens": 3916339.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.09182596206665,
|
|
"epoch": 1.9161772557394554,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004900584000872655,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.11126209571957588,
|
|
"num_tokens": 3927131.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.981594228744507,
|
|
"epoch": 1.9215162840363054,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004899337417803551,
|
|
"loss": 6.8705,
|
|
"mean_token_accuracy": 0.11410257518291474,
|
|
"num_tokens": 3938415.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.157159852981567,
|
|
"epoch": 1.9268553123331553,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000489808324665847,
|
|
"loss": 6.8829,
|
|
"mean_token_accuracy": 0.11425913646817207,
|
|
"num_tokens": 3949700.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.0690977573394775,
|
|
"epoch": 1.9321943406300055,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004896821491865167,
|
|
"loss": 6.9192,
|
|
"mean_token_accuracy": 0.10939718559384345,
|
|
"num_tokens": 3961879.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.1183741092681885,
|
|
"epoch": 1.9375333689268555,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004895552157878174,
|
|
"loss": 6.8921,
|
|
"mean_token_accuracy": 0.1135055162012577,
|
|
"num_tokens": 3972903.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.02628173828125,
|
|
"epoch": 1.9428723972237054,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004894275249178782,
|
|
"loss": 6.8852,
|
|
"mean_token_accuracy": 0.11946139633655548,
|
|
"num_tokens": 3984432.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.089619779586792,
|
|
"epoch": 1.9482114255205554,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004892990770275021,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.10337705612182617,
|
|
"num_tokens": 3996998.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.125215005874634,
|
|
"epoch": 1.9535504538174053,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004891698725701648,
|
|
"loss": 6.8477,
|
|
"mean_token_accuracy": 0.11870014742016792,
|
|
"num_tokens": 4007614.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.0385946273803714,
|
|
"epoch": 1.9588894821142553,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004890399120020131,
|
|
"loss": 6.8262,
|
|
"mean_token_accuracy": 0.10775638073682785,
|
|
"num_tokens": 4017645.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.1667829036712645,
|
|
"epoch": 1.9642285104111052,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004889091957818631,
|
|
"loss": 6.9244,
|
|
"mean_token_accuracy": 0.10332238003611564,
|
|
"num_tokens": 4029353.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.053739356994629,
|
|
"epoch": 1.9695675387079552,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004887777243711988,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.11049497202038765,
|
|
"num_tokens": 4040421.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.0194093704223635,
|
|
"epoch": 1.9749065670048052,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048864549823417,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.1106531098484993,
|
|
"num_tokens": 4051858.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.106148815155029,
|
|
"epoch": 1.9802455953016551,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004885125178375915,
|
|
"loss": 6.8803,
|
|
"mean_token_accuracy": 0.1077659197151661,
|
|
"num_tokens": 4062063.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.0133593559265135,
|
|
"epoch": 1.985584623598505,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004883787836509405,
|
|
"loss": 6.6985,
|
|
"mean_token_accuracy": 0.11615452989935875,
|
|
"num_tokens": 4072479.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.0427937507629395,
|
|
"epoch": 1.990923651895355,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004882442961463558,
|
|
"loss": 6.8677,
|
|
"mean_token_accuracy": 0.11139330714941025,
|
|
"num_tokens": 4082912.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.077240419387818,
|
|
"epoch": 1.996262680192205,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004881090557986353,
|
|
"loss": 6.8434,
|
|
"mean_token_accuracy": 0.11269779428839684,
|
|
"num_tokens": 4093644.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.915128972795275,
|
|
"epoch": 2.00106780565937,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004879730630852353,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.11212469720178181,
|
|
"num_tokens": 4104114.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.109417867660523,
|
|
"epoch": 2.0064068339562198,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048783631848626764,
|
|
"loss": 6.6362,
|
|
"mean_token_accuracy": 0.1157489076256752,
|
|
"num_tokens": 4114847.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.914434337615967,
|
|
"epoch": 2.0117458622530697,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004876988224844992,
|
|
"loss": 6.4861,
|
|
"mean_token_accuracy": 0.12341470643877983,
|
|
"num_tokens": 4126078.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.015130233764649,
|
|
"epoch": 2.01708489054992,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048756057556534934,
|
|
"loss": 6.5112,
|
|
"mean_token_accuracy": 0.12700850293040275,
|
|
"num_tokens": 4136678.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.966462326049805,
|
|
"epoch": 2.02242391884677,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004874215782168884,
|
|
"loss": 6.5355,
|
|
"mean_token_accuracy": 0.1265234611928463,
|
|
"num_tokens": 4146794.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.0654607772827145,
|
|
"epoch": 2.02776294714362,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004872818309298365,
|
|
"loss": 6.571,
|
|
"mean_token_accuracy": 0.12405075505375862,
|
|
"num_tokens": 4156825.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.920158243179321,
|
|
"epoch": 2.03310197544047,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048714133419756065,
|
|
"loss": 6.5027,
|
|
"mean_token_accuracy": 0.11929825246334076,
|
|
"num_tokens": 4166765.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.046618700027466,
|
|
"epoch": 2.03844100373732,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048700008851607437,
|
|
"loss": 6.6338,
|
|
"mean_token_accuracy": 0.12172048911452293,
|
|
"num_tokens": 4177404.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.052170658111573,
|
|
"epoch": 2.04378003203417,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.00048685809438403494,
|
|
"loss": 6.5971,
|
|
"mean_token_accuracy": 0.12037592306733132,
|
|
"num_tokens": 4188804.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.9027196884155275,
|
|
"epoch": 2.04911906033102,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004867153523027422,
|
|
"loss": 6.4989,
|
|
"mean_token_accuracy": 0.1245061419904232,
|
|
"num_tokens": 4200009.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.009029340744019,
|
|
"epoch": 2.05445808862787,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004865718627761363,
|
|
"loss": 6.5549,
|
|
"mean_token_accuracy": 0.12319564968347549,
|
|
"num_tokens": 4210776.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.948365020751953,
|
|
"epoch": 2.05979711692472,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048642762631079646,
|
|
"loss": 6.5094,
|
|
"mean_token_accuracy": 0.13088932484388352,
|
|
"num_tokens": 4221412.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.867213582992553,
|
|
"epoch": 2.0651361452215697,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004862826434159388,
|
|
"loss": 6.5012,
|
|
"mean_token_accuracy": 0.12945465818047525,
|
|
"num_tokens": 4231747.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.948292064666748,
|
|
"epoch": 2.0704751735184197,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048613691460341456,
|
|
"loss": 6.5181,
|
|
"mean_token_accuracy": 0.12791630178689956,
|
|
"num_tokens": 4241942.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.955828285217285,
|
|
"epoch": 2.0758142018152697,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00048599044038770856,
|
|
"loss": 6.6039,
|
|
"mean_token_accuracy": 0.12196211889386177,
|
|
"num_tokens": 4253151.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.93486852645874,
|
|
"epoch": 2.0811532301121196,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004858432212859372,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.12614453956484795,
|
|
"num_tokens": 4263579.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.9105040550231935,
|
|
"epoch": 2.0864922584089696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004856952578178464,
|
|
"loss": 6.5513,
|
|
"mean_token_accuracy": 0.12219369858503341,
|
|
"num_tokens": 4274764.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.026956415176391,
|
|
"epoch": 2.0918312867058195,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004855465505058102,
|
|
"loss": 6.6513,
|
|
"mean_token_accuracy": 0.11608931571245193,
|
|
"num_tokens": 4286561.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.919632291793823,
|
|
"epoch": 2.0971703150026695,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004853970998748289,
|
|
"loss": 6.603,
|
|
"mean_token_accuracy": 0.1245616041123867,
|
|
"num_tokens": 4296101.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.937914419174194,
|
|
"epoch": 2.1025093432995194,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048524690645252675,
|
|
"loss": 6.5519,
|
|
"mean_token_accuracy": 0.11856919229030609,
|
|
"num_tokens": 4307613.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.950258350372314,
|
|
"epoch": 2.1078483715963694,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004850959707691503,
|
|
"loss": 6.597,
|
|
"mean_token_accuracy": 0.12681118920445442,
|
|
"num_tokens": 4318042.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.949284172058105,
|
|
"epoch": 2.1131873998932194,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000484944293357567,
|
|
"loss": 6.4691,
|
|
"mean_token_accuracy": 0.12812729254364968,
|
|
"num_tokens": 4328690.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.885101842880249,
|
|
"epoch": 2.1185264281900693,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048479187475326254,
|
|
"loss": 6.5129,
|
|
"mean_token_accuracy": 0.1259954057633877,
|
|
"num_tokens": 4339188.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.827868366241455,
|
|
"epoch": 2.1238654564869193,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048463871549433947,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.1304394595324993,
|
|
"num_tokens": 4350242.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.834375524520874,
|
|
"epoch": 2.1292044847837692,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00048448481612151524,
|
|
"loss": 6.466,
|
|
"mean_token_accuracy": 0.13104819133877754,
|
|
"num_tokens": 4361138.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.958746910095215,
|
|
"epoch": 2.134543513080619,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048433017717812015,
|
|
"loss": 6.5523,
|
|
"mean_token_accuracy": 0.1202632024884224,
|
|
"num_tokens": 4372157.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.134543513080619,
|
|
"eval_entropy": 6.773753176107035,
|
|
"eval_loss": 6.934739112854004,
|
|
"eval_mean_token_accuracy": 0.11093968841936681,
|
|
"eval_num_tokens": 4372157.0,
|
|
"eval_runtime": 1.8151,
|
|
"eval_samples_per_second": 1696.873,
|
|
"eval_steps_per_second": 212.109,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.880448961257935,
|
|
"epoch": 2.139882541377469,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004841747992100954,
|
|
"loss": 6.5359,
|
|
"mean_token_accuracy": 0.12958219572901725,
|
|
"num_tokens": 4382735.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.900627279281617,
|
|
"epoch": 2.145221569674319,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048401868276599144,
|
|
"loss": 6.5906,
|
|
"mean_token_accuracy": 0.11768823489546776,
|
|
"num_tokens": 4393839.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.905943441390991,
|
|
"epoch": 2.150560597971169,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048386182839696575,
|
|
"loss": 6.5689,
|
|
"mean_token_accuracy": 0.12359676659107208,
|
|
"num_tokens": 4404612.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.915348386764526,
|
|
"epoch": 2.1558996262680195,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004837042366567809,
|
|
"loss": 6.5583,
|
|
"mean_token_accuracy": 0.12159886509180069,
|
|
"num_tokens": 4415969.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.947312211990356,
|
|
"epoch": 2.161238654564869,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048354590810180306,
|
|
"loss": 6.6298,
|
|
"mean_token_accuracy": 0.12084809169173241,
|
|
"num_tokens": 4427507.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.882000398635864,
|
|
"epoch": 2.1665776828617194,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004833868432909992,
|
|
"loss": 6.5736,
|
|
"mean_token_accuracy": 0.12455990388989449,
|
|
"num_tokens": 4438425.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.8933765411376955,
|
|
"epoch": 2.1719167111585693,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048322704278593595,
|
|
"loss": 6.5373,
|
|
"mean_token_accuracy": 0.1295296534895897,
|
|
"num_tokens": 4448301.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.85424017906189,
|
|
"epoch": 2.1772557394554193,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048306650715077703,
|
|
"loss": 6.5551,
|
|
"mean_token_accuracy": 0.12878405377268792,
|
|
"num_tokens": 4459214.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.8902589797973635,
|
|
"epoch": 2.1825947677522692,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048290523695228163,
|
|
"loss": 6.542,
|
|
"mean_token_accuracy": 0.1292489416897297,
|
|
"num_tokens": 4469696.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.856093168258667,
|
|
"epoch": 2.187933796049119,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004827432327598021,
|
|
"loss": 6.5611,
|
|
"mean_token_accuracy": 0.12441405653953552,
|
|
"num_tokens": 4481020.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.89211859703064,
|
|
"epoch": 2.193272824345969,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004825804951452823,
|
|
"loss": 6.5913,
|
|
"mean_token_accuracy": 0.115212332457304,
|
|
"num_tokens": 4492877.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.875864267349243,
|
|
"epoch": 2.198611852642819,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004824170246832553,
|
|
"loss": 6.5649,
|
|
"mean_token_accuracy": 0.12695193216204642,
|
|
"num_tokens": 4504184.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.8207022190094,
|
|
"epoch": 2.203950880939669,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 6.531,
|
|
"mean_token_accuracy": 0.12651038020849228,
|
|
"num_tokens": 4514807.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.828022575378418,
|
|
"epoch": 2.209289909236519,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048208788752774587,
|
|
"loss": 6.5443,
|
|
"mean_token_accuracy": 0.12110874876379966,
|
|
"num_tokens": 4526047.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.911036157608033,
|
|
"epoch": 2.214628937533369,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048192222199625783,
|
|
"loss": 6.4733,
|
|
"mean_token_accuracy": 0.13001587241888046,
|
|
"num_tokens": 4536506.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.8622376918792725,
|
|
"epoch": 2.219967965830219,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004817558259412469,
|
|
"loss": 6.6217,
|
|
"mean_token_accuracy": 0.12201437205076218,
|
|
"num_tokens": 4547351.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.931761169433594,
|
|
"epoch": 2.225306994127069,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004815886999501621,
|
|
"loss": 6.5862,
|
|
"mean_token_accuracy": 0.11850906684994697,
|
|
"num_tokens": 4558213.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.9577799320220945,
|
|
"epoch": 2.230646022423919,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004814208446130292,
|
|
"loss": 6.5277,
|
|
"mean_token_accuracy": 0.12626541927456855,
|
|
"num_tokens": 4567943.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.7742188453674315,
|
|
"epoch": 2.235985050720769,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048125226052244904,
|
|
"loss": 6.577,
|
|
"mean_token_accuracy": 0.12711381241679193,
|
|
"num_tokens": 4580000.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.921990203857422,
|
|
"epoch": 2.2413240790176188,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004810829482735952,
|
|
"loss": 6.5967,
|
|
"mean_token_accuracy": 0.12330932468175888,
|
|
"num_tokens": 4592160.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.837577629089355,
|
|
"epoch": 2.2466631073144687,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004809129084642121,
|
|
"loss": 6.5491,
|
|
"mean_token_accuracy": 0.1259201429784298,
|
|
"num_tokens": 4603747.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.867647123336792,
|
|
"epoch": 2.2520021356113187,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004807421416946125,
|
|
"loss": 6.5647,
|
|
"mean_token_accuracy": 0.126456468552351,
|
|
"num_tokens": 4615660.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.81969141960144,
|
|
"epoch": 2.2573411639081686,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004805706485676759,
|
|
"loss": 6.575,
|
|
"mean_token_accuracy": 0.12571234479546547,
|
|
"num_tokens": 4626408.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.942286539077759,
|
|
"epoch": 2.2626801922050186,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004803984296888459,
|
|
"loss": 6.5099,
|
|
"mean_token_accuracy": 0.12510694563388824,
|
|
"num_tokens": 4638236.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.795414113998413,
|
|
"epoch": 2.2680192205018685,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048022548566612855,
|
|
"loss": 6.5967,
|
|
"mean_token_accuracy": 0.12317438274621964,
|
|
"num_tokens": 4649649.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.934534072875977,
|
|
"epoch": 2.2733582487987185,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004800518171100899,
|
|
"loss": 6.5907,
|
|
"mean_token_accuracy": 0.12582072243094444,
|
|
"num_tokens": 4660502.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.865882444381714,
|
|
"epoch": 2.2786972770955685,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004798774246338538,
|
|
"loss": 6.5365,
|
|
"mean_token_accuracy": 0.125477909296751,
|
|
"num_tokens": 4671486.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.848501825332642,
|
|
"epoch": 2.2840363053924184,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004797023088530999,
|
|
"loss": 6.5941,
|
|
"mean_token_accuracy": 0.125645562261343,
|
|
"num_tokens": 4681436.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.98210334777832,
|
|
"epoch": 2.2893753336892684,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047952647038606157,
|
|
"loss": 6.5881,
|
|
"mean_token_accuracy": 0.1284101627767086,
|
|
"num_tokens": 4692065.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.836091232299805,
|
|
"epoch": 2.2947143619861183,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004793499098535233,
|
|
"loss": 6.5034,
|
|
"mean_token_accuracy": 0.1249152347445488,
|
|
"num_tokens": 4702107.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.831265258789062,
|
|
"epoch": 2.3000533902829687,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004791726278788191,
|
|
"loss": 6.6077,
|
|
"mean_token_accuracy": 0.12131517902016639,
|
|
"num_tokens": 4712060.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.904486179351807,
|
|
"epoch": 2.3053924185798182,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004789946250878296,
|
|
"loss": 6.5162,
|
|
"mean_token_accuracy": 0.13010889962315558,
|
|
"num_tokens": 4721805.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.740131425857544,
|
|
"epoch": 2.3107314468766686,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004788159021089805,
|
|
"loss": 6.5035,
|
|
"mean_token_accuracy": 0.1257294774055481,
|
|
"num_tokens": 4733630.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.8522107124328615,
|
|
"epoch": 2.3160704751735186,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004786364595732401,
|
|
"loss": 6.5728,
|
|
"mean_token_accuracy": 0.12956593036651612,
|
|
"num_tokens": 4743634.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.872754287719727,
|
|
"epoch": 2.3214095034703686,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004784562981141168,
|
|
"loss": 6.5247,
|
|
"mean_token_accuracy": 0.12224958017468453,
|
|
"num_tokens": 4754815.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.810030364990235,
|
|
"epoch": 2.3267485317672185,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004782754183676573,
|
|
"loss": 6.5851,
|
|
"mean_token_accuracy": 0.12206983789801598,
|
|
"num_tokens": 4765657.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.9540458679199215,
|
|
"epoch": 2.3320875600640685,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00047809382097244407,
|
|
"loss": 6.6016,
|
|
"mean_token_accuracy": 0.12477592378854752,
|
|
"num_tokens": 4777436.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.874520349502563,
|
|
"epoch": 2.3374265883609184,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047791150656959326,
|
|
"loss": 6.587,
|
|
"mean_token_accuracy": 0.12307127118110657,
|
|
"num_tokens": 4788149.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.784716367721558,
|
|
"epoch": 2.3427656166577684,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047772847580275225,
|
|
"loss": 6.5332,
|
|
"mean_token_accuracy": 0.12545752599835397,
|
|
"num_tokens": 4799512.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.875580453872681,
|
|
"epoch": 2.3481046449546183,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004775447293180976,
|
|
"loss": 6.5423,
|
|
"mean_token_accuracy": 0.12484646216034889,
|
|
"num_tokens": 4810410.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.906468296051026,
|
|
"epoch": 2.3534436732514683,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00047736026776433254,
|
|
"loss": 6.6885,
|
|
"mean_token_accuracy": 0.11681632474064826,
|
|
"num_tokens": 4822247.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.934691524505615,
|
|
"epoch": 2.3587827015483183,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047717509179268504,
|
|
"loss": 6.555,
|
|
"mean_token_accuracy": 0.1287574701011181,
|
|
"num_tokens": 4833550.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.831350469589234,
|
|
"epoch": 2.364121729845168,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00047698920205690494,
|
|
"loss": 6.5855,
|
|
"mean_token_accuracy": 0.12659201845526696,
|
|
"num_tokens": 4845095.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.821032953262329,
|
|
"epoch": 2.369460758142018,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047680259921326225,
|
|
"loss": 6.504,
|
|
"mean_token_accuracy": 0.12526951134204864,
|
|
"num_tokens": 4856641.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.866345167160034,
|
|
"epoch": 2.374799786438868,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004766152839205443,
|
|
"loss": 6.5542,
|
|
"mean_token_accuracy": 0.12299052476882935,
|
|
"num_tokens": 4868274.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.865297317504883,
|
|
"epoch": 2.380138814735718,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000476427256840054,
|
|
"loss": 6.5413,
|
|
"mean_token_accuracy": 0.12318603098392486,
|
|
"num_tokens": 4878308.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.845654821395874,
|
|
"epoch": 2.385477843032568,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004762385186356068,
|
|
"loss": 6.5951,
|
|
"mean_token_accuracy": 0.12697439640760422,
|
|
"num_tokens": 4888811.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.8985631465911865,
|
|
"epoch": 2.390816871329418,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004760490699735289,
|
|
"loss": 6.5085,
|
|
"mean_token_accuracy": 0.13273092210292817,
|
|
"num_tokens": 4898719.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.801689481735229,
|
|
"epoch": 2.396155899626268,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004758589115226549,
|
|
"loss": 6.5474,
|
|
"mean_token_accuracy": 0.12448980137705803,
|
|
"num_tokens": 4909480.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.8546003818511965,
|
|
"epoch": 2.401494927923118,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047566804395432493,
|
|
"loss": 6.5932,
|
|
"mean_token_accuracy": 0.1219810850918293,
|
|
"num_tokens": 4920560.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.816539764404297,
|
|
"epoch": 2.406833956219968,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 6.4552,
|
|
"mean_token_accuracy": 0.12189477160573006,
|
|
"num_tokens": 4931499.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.754684591293335,
|
|
"epoch": 2.412172984516818,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047528418416317334,
|
|
"loss": 6.5083,
|
|
"mean_token_accuracy": 0.13347715884447098,
|
|
"num_tokens": 4942133.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.8414233207702635,
|
|
"epoch": 2.417512012813668,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004750911932955402,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.12275219932198525,
|
|
"num_tokens": 4953255.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.86159462928772,
|
|
"epoch": 2.4228510411105177,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047489749602082316,
|
|
"loss": 6.5256,
|
|
"mean_token_accuracy": 0.12575036883354188,
|
|
"num_tokens": 4964766.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.822673892974853,
|
|
"epoch": 2.4281900694073677,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047470309302285617,
|
|
"loss": 6.5921,
|
|
"mean_token_accuracy": 0.12224113792181016,
|
|
"num_tokens": 4976644.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.7894798755645756,
|
|
"epoch": 2.4335290977042177,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047450798498796466,
|
|
"loss": 6.5303,
|
|
"mean_token_accuracy": 0.12349486351013184,
|
|
"num_tokens": 4986976.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.881432390213012,
|
|
"epoch": 2.4388681260010676,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004743121726049628,
|
|
"loss": 6.5324,
|
|
"mean_token_accuracy": 0.12652337923645973,
|
|
"num_tokens": 4998747.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.814818048477173,
|
|
"epoch": 2.444207154297918,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004741156565651518,
|
|
"loss": 6.5198,
|
|
"mean_token_accuracy": 0.13365631699562072,
|
|
"num_tokens": 5009323.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.901062488555908,
|
|
"epoch": 2.4495461825947675,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047391843756231693,
|
|
"loss": 6.5853,
|
|
"mean_token_accuracy": 0.12531132400035858,
|
|
"num_tokens": 5019434.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.782677125930786,
|
|
"epoch": 2.454885210891618,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00047372051629272516,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.1259647473692894,
|
|
"num_tokens": 5029884.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.791875982284546,
|
|
"epoch": 2.460224239188468,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047352189345512284,
|
|
"loss": 6.4898,
|
|
"mean_token_accuracy": 0.13430592641234398,
|
|
"num_tokens": 5040323.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.763852119445801,
|
|
"epoch": 2.465563267485318,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004733225697507332,
|
|
"loss": 6.5126,
|
|
"mean_token_accuracy": 0.12424452975392342,
|
|
"num_tokens": 5051363.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.80372052192688,
|
|
"epoch": 2.470902295782168,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004731225458832537,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.13885923177003862,
|
|
"num_tokens": 5061221.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.77863450050354,
|
|
"epoch": 2.4762413240790178,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00047292182255885367,
|
|
"loss": 6.4959,
|
|
"mean_token_accuracy": 0.13018832132220268,
|
|
"num_tokens": 5072578.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.769831752777099,
|
|
"epoch": 2.4815803523758677,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004727204004861721,
|
|
"loss": 6.4961,
|
|
"mean_token_accuracy": 0.12298517078161239,
|
|
"num_tokens": 5083244.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.7560343742370605,
|
|
"epoch": 2.4869193806727177,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004725182803763144,
|
|
"loss": 6.5339,
|
|
"mean_token_accuracy": 0.1289779581129551,
|
|
"num_tokens": 5094115.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.829773807525635,
|
|
"epoch": 2.4922584089695676,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004723154629428507,
|
|
"loss": 6.5461,
|
|
"mean_token_accuracy": 0.12436275333166122,
|
|
"num_tokens": 5104838.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.859137296676636,
|
|
"epoch": 2.4975974372664176,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004721119489018128,
|
|
"loss": 6.5327,
|
|
"mean_token_accuracy": 0.12045159563422203,
|
|
"num_tokens": 5115078.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.797474527359009,
|
|
"epoch": 2.5029364655632675,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047190773897169203,
|
|
"loss": 6.5572,
|
|
"mean_token_accuracy": 0.12596193179488183,
|
|
"num_tokens": 5126414.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.828856039047241,
|
|
"epoch": 2.5082754938601175,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047170283387343634,
|
|
"loss": 6.562,
|
|
"mean_token_accuracy": 0.12051519602537156,
|
|
"num_tokens": 5136271.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.800413131713867,
|
|
"epoch": 2.5136145221569675,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047149723433044787,
|
|
"loss": 6.567,
|
|
"mean_token_accuracy": 0.12625120356678962,
|
|
"num_tokens": 5146780.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.86309061050415,
|
|
"epoch": 2.5189535504538174,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047129094106858056,
|
|
"loss": 6.5779,
|
|
"mean_token_accuracy": 0.12552282735705375,
|
|
"num_tokens": 5157941.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.79384446144104,
|
|
"epoch": 2.5242925787506674,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047108395481613736,
|
|
"loss": 6.5004,
|
|
"mean_token_accuracy": 0.1237537682056427,
|
|
"num_tokens": 5168994.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.758298110961914,
|
|
"epoch": 2.5296316070475173,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00047087627630386803,
|
|
"loss": 6.4828,
|
|
"mean_token_accuracy": 0.1223898023366928,
|
|
"num_tokens": 5180780.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.754440212249756,
|
|
"epoch": 2.5349706353443673,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000470667906264966,
|
|
"loss": 6.5634,
|
|
"mean_token_accuracy": 0.12323177009820938,
|
|
"num_tokens": 5192090.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.85933289527893,
|
|
"epoch": 2.5403096636412172,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004704588454350661,
|
|
"loss": 6.4706,
|
|
"mean_token_accuracy": 0.1317416600883007,
|
|
"num_tokens": 5201854.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.773879671096802,
|
|
"epoch": 2.545648691938067,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004702490945522423,
|
|
"loss": 6.4941,
|
|
"mean_token_accuracy": 0.13044386208057404,
|
|
"num_tokens": 5213998.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.708115959167481,
|
|
"epoch": 2.550987720234917,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004700386543570043,
|
|
"loss": 6.4527,
|
|
"mean_token_accuracy": 0.12814613208174705,
|
|
"num_tokens": 5225217.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.704124879837036,
|
|
"epoch": 2.556326748531767,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004698275255922956,
|
|
"loss": 6.3849,
|
|
"mean_token_accuracy": 0.1327833689749241,
|
|
"num_tokens": 5235098.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.790454530715943,
|
|
"epoch": 2.561665776828617,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046961570900349073,
|
|
"loss": 6.5778,
|
|
"mean_token_accuracy": 0.12630253434181213,
|
|
"num_tokens": 5246704.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.754661941528321,
|
|
"epoch": 2.567004805125467,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004694032053383922,
|
|
"loss": 6.4673,
|
|
"mean_token_accuracy": 0.13032191917300223,
|
|
"num_tokens": 5257331.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.766871213912964,
|
|
"epoch": 2.572343833422317,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004691900153472286,
|
|
"loss": 6.4521,
|
|
"mean_token_accuracy": 0.13126400411128997,
|
|
"num_tokens": 5267662.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.807767772674561,
|
|
"epoch": 2.5776828617191674,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046897613978265124,
|
|
"loss": 6.4927,
|
|
"mean_token_accuracy": 0.12516679093241692,
|
|
"num_tokens": 5277977.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.830004692077637,
|
|
"epoch": 2.583021890016017,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004687615793997319,
|
|
"loss": 6.6179,
|
|
"mean_token_accuracy": 0.11929458528757095,
|
|
"num_tokens": 5288987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.791280221939087,
|
|
"epoch": 2.5883609183128673,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004685463349559602,
|
|
"loss": 6.4433,
|
|
"mean_token_accuracy": 0.12909359782934188,
|
|
"num_tokens": 5300300.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.760449028015136,
|
|
"epoch": 2.593699946609717,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00046833040721124035,
|
|
"loss": 6.464,
|
|
"mean_token_accuracy": 0.12651569321751593,
|
|
"num_tokens": 5311606.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.779611396789551,
|
|
"epoch": 2.599038974906567,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004681137969278896,
|
|
"loss": 6.608,
|
|
"mean_token_accuracy": 0.11662786602973937,
|
|
"num_tokens": 5323218.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.729658365249634,
|
|
"epoch": 2.6043780032034167,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004678965048706341,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.13026198297739028,
|
|
"num_tokens": 5334516.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.765009355545044,
|
|
"epoch": 2.609717031500267,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 6.5194,
|
|
"mean_token_accuracy": 0.12988420128822326,
|
|
"num_tokens": 5345751.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.8113123893737795,
|
|
"epoch": 2.615056059797117,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004674598785053475,
|
|
"loss": 6.5869,
|
|
"mean_token_accuracy": 0.12195999845862389,
|
|
"num_tokens": 5356918.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.801035404205322,
|
|
"epoch": 2.620395088093967,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046724054573879314,
|
|
"loss": 6.4824,
|
|
"mean_token_accuracy": 0.13370697051286698,
|
|
"num_tokens": 5368175.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.681986713409424,
|
|
"epoch": 2.625734116390817,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004670205342812827,
|
|
"loss": 6.4613,
|
|
"mean_token_accuracy": 0.128626449406147,
|
|
"num_tokens": 5379074.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.836401033401489,
|
|
"epoch": 2.631073144687667,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00046679984490955015,
|
|
"loss": 6.5203,
|
|
"mean_token_accuracy": 0.12719035670161247,
|
|
"num_tokens": 5389380.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.695728015899658,
|
|
"epoch": 2.636412172984517,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004665784784027229,
|
|
"loss": 6.5465,
|
|
"mean_token_accuracy": 0.1358778417110443,
|
|
"num_tokens": 5401378.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.82390513420105,
|
|
"epoch": 2.641751201281367,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004663564355423189,
|
|
"loss": 6.4823,
|
|
"mean_token_accuracy": 0.12906705513596534,
|
|
"num_tokens": 5411434.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.819698429107666,
|
|
"epoch": 2.647090229578217,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004661337171122442,
|
|
"loss": 6.4761,
|
|
"mean_token_accuracy": 0.1277168557047844,
|
|
"num_tokens": 5421597.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.689829397201538,
|
|
"epoch": 2.652429257875067,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00046591032389878934,
|
|
"loss": 6.485,
|
|
"mean_token_accuracy": 0.126693993806839,
|
|
"num_tokens": 5432166.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.698560667037964,
|
|
"epoch": 2.6577682861719167,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004656862566906276,
|
|
"loss": 6.4949,
|
|
"mean_token_accuracy": 0.12689822018146515,
|
|
"num_tokens": 5443156.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.902831506729126,
|
|
"epoch": 2.6631073144687667,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046546151627881166,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.11838782355189323,
|
|
"num_tokens": 5453830.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.678612470626831,
|
|
"epoch": 2.6684463427656167,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004652361034567708,
|
|
"loss": 6.4709,
|
|
"mean_token_accuracy": 0.1326916351914406,
|
|
"num_tokens": 5465371.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.6684463427656167,
|
|
"eval_entropy": 6.506537835009686,
|
|
"eval_loss": 6.804722785949707,
|
|
"eval_mean_token_accuracy": 0.11863555662430726,
|
|
"eval_num_tokens": 5465371.0,
|
|
"eval_runtime": 1.8173,
|
|
"eval_samples_per_second": 1694.823,
|
|
"eval_steps_per_second": 211.853,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.768008804321289,
|
|
"epoch": 2.6737853710624666,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046501001902030814,
|
|
"loss": 6.5267,
|
|
"mean_token_accuracy": 0.1259946919977665,
|
|
"num_tokens": 5476687.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.789381408691407,
|
|
"epoch": 2.6791243993593166,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004647832637675982,
|
|
"loss": 6.4999,
|
|
"mean_token_accuracy": 0.1216110959649086,
|
|
"num_tokens": 5487825.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.750199174880981,
|
|
"epoch": 2.6844634276561665,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046455583849918324,
|
|
"loss": 6.4768,
|
|
"mean_token_accuracy": 0.13176634460687636,
|
|
"num_tokens": 5498561.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.7578329086303714,
|
|
"epoch": 2.6898024559530165,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004643277440179715,
|
|
"loss": 6.4973,
|
|
"mean_token_accuracy": 0.12797005027532576,
|
|
"num_tokens": 5509624.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.720186614990235,
|
|
"epoch": 2.6951414842498664,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004640989811292335,
|
|
"loss": 6.3996,
|
|
"mean_token_accuracy": 0.13824927657842637,
|
|
"num_tokens": 5519484.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.689453601837158,
|
|
"epoch": 2.7004805125467164,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004638695506405996,
|
|
"loss": 6.3951,
|
|
"mean_token_accuracy": 0.132322496175766,
|
|
"num_tokens": 5531839.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.736231422424316,
|
|
"epoch": 2.7058195408435664,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004636394533620571,
|
|
"loss": 6.5275,
|
|
"mean_token_accuracy": 0.1268465995788574,
|
|
"num_tokens": 5541787.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.809881973266601,
|
|
"epoch": 2.7111585691404163,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004634086901059474,
|
|
"loss": 6.582,
|
|
"mean_token_accuracy": 0.1202095814049244,
|
|
"num_tokens": 5553493.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.805474138259887,
|
|
"epoch": 2.7164975974372663,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004631772616869629,
|
|
"loss": 6.4382,
|
|
"mean_token_accuracy": 0.12904566749930382,
|
|
"num_tokens": 5564809.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.642514276504516,
|
|
"epoch": 2.7218366257341167,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00046294516892214454,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.12669157981872559,
|
|
"num_tokens": 5575575.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.770498991012573,
|
|
"epoch": 2.727175654030966,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00046271241263087846,
|
|
"loss": 6.4719,
|
|
"mean_token_accuracy": 0.12543186768889428,
|
|
"num_tokens": 5587276.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.732878065109253,
|
|
"epoch": 2.7325146823278166,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004624789936348936,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.1274913065135479,
|
|
"num_tokens": 5597557.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.679828453063965,
|
|
"epoch": 2.737853710624666,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046224491275825825,
|
|
"loss": 6.4392,
|
|
"mean_token_accuracy": 0.13149973079562188,
|
|
"num_tokens": 5608512.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.733901882171631,
|
|
"epoch": 2.7431927389215165,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004620101708273776,
|
|
"loss": 6.5047,
|
|
"mean_token_accuracy": 0.13109129667282104,
|
|
"num_tokens": 5621788.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.806593799591065,
|
|
"epoch": 2.748531767218366,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046177476867099055,
|
|
"loss": 6.4933,
|
|
"mean_token_accuracy": 0.12798713520169258,
|
|
"num_tokens": 5632493.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.739014434814453,
|
|
"epoch": 2.7538707955152164,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046153870712016694,
|
|
"loss": 6.5157,
|
|
"mean_token_accuracy": 0.12476658746600151,
|
|
"num_tokens": 5643163.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.717135906219482,
|
|
"epoch": 2.7592098238120664,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004613019870083045,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.1334136761724949,
|
|
"num_tokens": 5653362.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.775650072097778,
|
|
"epoch": 2.7645488521089163,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004610646091711261,
|
|
"loss": 6.5442,
|
|
"mean_token_accuracy": 0.13016111105680467,
|
|
"num_tokens": 5663637.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.6957603931427006,
|
|
"epoch": 2.7698878804057663,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004608265744466763,
|
|
"loss": 6.4834,
|
|
"mean_token_accuracy": 0.1269044943153858,
|
|
"num_tokens": 5674446.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.8000424861907955,
|
|
"epoch": 2.7752269087026162,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046058788367531927,
|
|
"loss": 6.5139,
|
|
"mean_token_accuracy": 0.13062895238399505,
|
|
"num_tokens": 5684768.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.765134954452515,
|
|
"epoch": 2.780565936999466,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004603485376997349,
|
|
"loss": 6.4089,
|
|
"mean_token_accuracy": 0.13543231412768364,
|
|
"num_tokens": 5696474.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.602987575531006,
|
|
"epoch": 2.785904965296316,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004601085373649163,
|
|
"loss": 6.5767,
|
|
"mean_token_accuracy": 0.12214590460062028,
|
|
"num_tokens": 5707829.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.819445943832397,
|
|
"epoch": 2.791243993593166,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000459867883518167,
|
|
"loss": 6.5069,
|
|
"mean_token_accuracy": 0.12737508937716485,
|
|
"num_tokens": 5719325.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.790378141403198,
|
|
"epoch": 2.796583021890016,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004596265770090974,
|
|
"loss": 6.4816,
|
|
"mean_token_accuracy": 0.12503273263573647,
|
|
"num_tokens": 5730859.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.709441614151001,
|
|
"epoch": 2.801922050186866,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004593846186896222,
|
|
"loss": 6.4519,
|
|
"mean_token_accuracy": 0.132213544100523,
|
|
"num_tokens": 5741595.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.777496957778931,
|
|
"epoch": 2.807261078483716,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00045914200941395725,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.12735849544405936,
|
|
"num_tokens": 5752989.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.731167650222778,
|
|
"epoch": 2.812600106780566,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004588987500386165,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.12772611528635025,
|
|
"num_tokens": 5764440.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.696313142776489,
|
|
"epoch": 2.817939135077416,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004586548414224093,
|
|
"loss": 6.3725,
|
|
"mean_token_accuracy": 0.13323593363165856,
|
|
"num_tokens": 5776389.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.725729465484619,
|
|
"epoch": 2.823278163374266,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045841028442643675,
|
|
"loss": 6.4058,
|
|
"mean_token_accuracy": 0.13175813257694244,
|
|
"num_tokens": 5787803.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.698521423339844,
|
|
"epoch": 2.828617191671116,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004581650799140891,
|
|
"loss": 6.4867,
|
|
"mean_token_accuracy": 0.12856226190924644,
|
|
"num_tokens": 5799647.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.737258195877075,
|
|
"epoch": 2.8339562199679658,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004579192287510427,
|
|
"loss": 6.4974,
|
|
"mean_token_accuracy": 0.12667600736021994,
|
|
"num_tokens": 5809968.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.778554439544678,
|
|
"epoch": 2.8392952482648157,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004576727318052568,
|
|
"loss": 6.465,
|
|
"mean_token_accuracy": 0.13686817660927772,
|
|
"num_tokens": 5820075.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.646070718765259,
|
|
"epoch": 2.8446342765616657,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045742558994697063,
|
|
"loss": 6.4838,
|
|
"mean_token_accuracy": 0.13021821826696395,
|
|
"num_tokens": 5830686.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.814020347595215,
|
|
"epoch": 2.8499733048585156,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004571778040487001,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.13131239861249924,
|
|
"num_tokens": 5841451.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.666039228439331,
|
|
"epoch": 2.8553123331553656,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004569293749852347,
|
|
"loss": 6.4986,
|
|
"mean_token_accuracy": 0.1276731461286545,
|
|
"num_tokens": 5852650.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.715946388244629,
|
|
"epoch": 2.8606513614522155,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004566803036336351,
|
|
"loss": 6.4434,
|
|
"mean_token_accuracy": 0.1310188926756382,
|
|
"num_tokens": 5865056.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.740395164489746,
|
|
"epoch": 2.865990389749066,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045643059087322895,
|
|
"loss": 6.4935,
|
|
"mean_token_accuracy": 0.12656164914369583,
|
|
"num_tokens": 5876069.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.742620897293091,
|
|
"epoch": 2.8713294180459155,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004561802375856088,
|
|
"loss": 6.4406,
|
|
"mean_token_accuracy": 0.1361336812376976,
|
|
"num_tokens": 5886366.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.720592260360718,
|
|
"epoch": 2.876668446342766,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004559292446546281,
|
|
"loss": 6.508,
|
|
"mean_token_accuracy": 0.1300479516386986,
|
|
"num_tokens": 5897101.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.787854623794556,
|
|
"epoch": 2.8820074746396154,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045567761296639887,
|
|
"loss": 6.5176,
|
|
"mean_token_accuracy": 0.1308740459382534,
|
|
"num_tokens": 5908748.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.783561897277832,
|
|
"epoch": 2.8873465029364658,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000455425343409288,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.12536448761820793,
|
|
"num_tokens": 5920830.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.626713418960572,
|
|
"epoch": 2.8926855312333153,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045517243687391454,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.13662005811929703,
|
|
"num_tokens": 5932360.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.75290880203247,
|
|
"epoch": 2.8980245595301657,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004549188942531461,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.12693905979394912,
|
|
"num_tokens": 5944067.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.806321859359741,
|
|
"epoch": 2.9033635878270156,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000454664716442096,
|
|
"loss": 6.5369,
|
|
"mean_token_accuracy": 0.1230388343334198,
|
|
"num_tokens": 5955974.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.792594861984253,
|
|
"epoch": 2.9087026161238656,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045440990433812025,
|
|
"loss": 6.5309,
|
|
"mean_token_accuracy": 0.1249440647661686,
|
|
"num_tokens": 5966542.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.660552597045898,
|
|
"epoch": 2.9140416444207156,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000454154458840814,
|
|
"loss": 6.4113,
|
|
"mean_token_accuracy": 0.1303204596042633,
|
|
"num_tokens": 5977342.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.760577392578125,
|
|
"epoch": 2.9193806727175655,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004538983808520085,
|
|
"loss": 6.5565,
|
|
"mean_token_accuracy": 0.12395976111292839,
|
|
"num_tokens": 5988786.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.788736200332641,
|
|
"epoch": 2.9247197010144155,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045364167127576817,
|
|
"loss": 6.5493,
|
|
"mean_token_accuracy": 0.11930835843086243,
|
|
"num_tokens": 6000167.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.671005630493164,
|
|
"epoch": 2.9300587293112654,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045338433101838705,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.12854471430182457,
|
|
"num_tokens": 6012214.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.764946031570434,
|
|
"epoch": 2.9353977576081154,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045312636098838576,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.13039105236530305,
|
|
"num_tokens": 6023032.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.726640939712524,
|
|
"epoch": 2.9407367859049653,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004528677620965083,
|
|
"loss": 6.4315,
|
|
"mean_token_accuracy": 0.13164386823773383,
|
|
"num_tokens": 6033398.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.7339318752288815,
|
|
"epoch": 2.9460758142018153,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045260853525571894,
|
|
"loss": 6.4657,
|
|
"mean_token_accuracy": 0.12935673370957373,
|
|
"num_tokens": 6044549.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.581913232803345,
|
|
"epoch": 2.9514148424986653,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004523486813811987,
|
|
"loss": 6.2531,
|
|
"mean_token_accuracy": 0.13347729817032813,
|
|
"num_tokens": 6055213.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.656859970092773,
|
|
"epoch": 2.956753870795515,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004520882013903423,
|
|
"loss": 6.3928,
|
|
"mean_token_accuracy": 0.13870861008763313,
|
|
"num_tokens": 6065488.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.702912044525147,
|
|
"epoch": 2.962092899092365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045182709620275504,
|
|
"loss": 6.4888,
|
|
"mean_token_accuracy": 0.1314636677503586,
|
|
"num_tokens": 6077525.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.749990224838257,
|
|
"epoch": 2.967431927389215,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004515653667402494,
|
|
"loss": 6.4758,
|
|
"mean_token_accuracy": 0.1329290971159935,
|
|
"num_tokens": 6088707.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.70402946472168,
|
|
"epoch": 2.972770955686065,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004513030139268418,
|
|
"loss": 6.4342,
|
|
"mean_token_accuracy": 0.13106571063399314,
|
|
"num_tokens": 6099231.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.698221158981323,
|
|
"epoch": 2.978109983982915,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004510400386887493,
|
|
"loss": 6.5101,
|
|
"mean_token_accuracy": 0.12947579026222228,
|
|
"num_tokens": 6110356.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.7977382183074955,
|
|
"epoch": 2.983449012279765,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045077644195438645,
|
|
"loss": 6.4675,
|
|
"mean_token_accuracy": 0.1286630518734455,
|
|
"num_tokens": 6120722.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.746077966690064,
|
|
"epoch": 2.988788040576615,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000450512224654362,
|
|
"loss": 6.4564,
|
|
"mean_token_accuracy": 0.13280715346336364,
|
|
"num_tokens": 6131404.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.678772449493408,
|
|
"epoch": 2.994127068873465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045024738772147534,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.12211451306939125,
|
|
"num_tokens": 6141402.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.752857732772827,
|
|
"epoch": 2.999466097170315,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00044998193209071365,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.13422416672110557,
|
|
"num_tokens": 6152082.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.670749982198079,
|
|
"epoch": 3.00427122263748,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00044971585869924826,
|
|
"loss": 6.124,
|
|
"mean_token_accuracy": 0.14834013250139025,
|
|
"num_tokens": 6162282.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.658770990371704,
|
|
"epoch": 3.00961025093433,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004494491684864316,
|
|
"loss": 6.1142,
|
|
"mean_token_accuracy": 0.13844860345125198,
|
|
"num_tokens": 6173818.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.700858926773071,
|
|
"epoch": 3.01494927923118,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 6.145,
|
|
"mean_token_accuracy": 0.1486111745238304,
|
|
"num_tokens": 6184406.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.702039670944214,
|
|
"epoch": 3.02028830752803,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004489139413650385,
|
|
"loss": 6.1982,
|
|
"mean_token_accuracy": 0.13922229036688805,
|
|
"num_tokens": 6195817.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.690669584274292,
|
|
"epoch": 3.02562733582488,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004486454063460415,
|
|
"loss": 6.0552,
|
|
"mean_token_accuracy": 0.14872528836131096,
|
|
"num_tokens": 6206176.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.633033227920532,
|
|
"epoch": 3.03096636412173,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004483762582848457,
|
|
"loss": 6.1384,
|
|
"mean_token_accuracy": 0.14411605522036552,
|
|
"num_tokens": 6215783.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.706024503707885,
|
|
"epoch": 3.03630539241858,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004481064981316583,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.13902800604701043,
|
|
"num_tokens": 6227146.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.64505181312561,
|
|
"epoch": 3.04164442071543,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00044783612683884744,
|
|
"loss": 6.1002,
|
|
"mean_token_accuracy": 0.14859205186367036,
|
|
"num_tokens": 6237417.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.639095830917358,
|
|
"epoch": 3.04698344901228,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004475651453609388,
|
|
"loss": 6.1527,
|
|
"mean_token_accuracy": 0.14703191369771956,
|
|
"num_tokens": 6249136.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.657324695587159,
|
|
"epoch": 3.0523224773091298,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00044729355465461236,
|
|
"loss": 6.1194,
|
|
"mean_token_accuracy": 0.1403007425367832,
|
|
"num_tokens": 6260811.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.6688212871551515,
|
|
"epoch": 3.0576615056059797,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000447021355678699,
|
|
"loss": 6.1916,
|
|
"mean_token_accuracy": 0.13820797055959702,
|
|
"num_tokens": 6272776.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.72321252822876,
|
|
"epoch": 3.0630005339028297,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004467485493941769,
|
|
"loss": 6.1175,
|
|
"mean_token_accuracy": 0.1424892447888851,
|
|
"num_tokens": 6283335.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.535885858535766,
|
|
"epoch": 3.0683395621996796,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044647513676416835,
|
|
"loss": 6.0816,
|
|
"mean_token_accuracy": 0.14990985095500947,
|
|
"num_tokens": 6293383.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.673948097229004,
|
|
"epoch": 3.0736785904965296,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004462011187539364,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.1540223926305771,
|
|
"num_tokens": 6303373.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.657982444763183,
|
|
"epoch": 3.0790176187933795,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004459264963308812,
|
|
"loss": 6.1452,
|
|
"mean_token_accuracy": 0.1427639067173004,
|
|
"num_tokens": 6315674.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.5796790599823,
|
|
"epoch": 3.0843566470902295,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044565127046453687,
|
|
"loss": 6.0308,
|
|
"mean_token_accuracy": 0.14728442281484605,
|
|
"num_tokens": 6326183.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.510620546340943,
|
|
"epoch": 3.0896956753870795,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00044537544212656775,
|
|
"loss": 6.1323,
|
|
"mean_token_accuracy": 0.14221204370260238,
|
|
"num_tokens": 6337495.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.6099615573883055,
|
|
"epoch": 3.0950347036839294,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00044509901229076537,
|
|
"loss": 6.0958,
|
|
"mean_token_accuracy": 0.14601162523031236,
|
|
"num_tokens": 6348935.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.617335224151612,
|
|
"epoch": 3.1003737319807794,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004448219819330447,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.13788999915122985,
|
|
"num_tokens": 6359701.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.688894605636596,
|
|
"epoch": 3.1057127602776293,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004445443520314409,
|
|
"loss": 6.1635,
|
|
"mean_token_accuracy": 0.1362752579152584,
|
|
"num_tokens": 6370134.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.596358346939087,
|
|
"epoch": 3.1110517885744793,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044426612356610555,
|
|
"loss": 6.1984,
|
|
"mean_token_accuracy": 0.1391242265701294,
|
|
"num_tokens": 6381644.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.677271556854248,
|
|
"epoch": 3.1163908168713292,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004439872975193037,
|
|
"loss": 6.1728,
|
|
"mean_token_accuracy": 0.1386138141155243,
|
|
"num_tokens": 6393386.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.570891571044922,
|
|
"epoch": 3.121729845168179,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00044370787487540986,
|
|
"loss": 6.1742,
|
|
"mean_token_accuracy": 0.14069369807839394,
|
|
"num_tokens": 6405487.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.6059774398803714,
|
|
"epoch": 3.1270688734650296,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000443427856620905,
|
|
"loss": 6.1768,
|
|
"mean_token_accuracy": 0.14093007072806357,
|
|
"num_tokens": 6416470.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.669240856170655,
|
|
"epoch": 3.132407901761879,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044314724374437274,
|
|
"loss": 6.2054,
|
|
"mean_token_accuracy": 0.14186814278364182,
|
|
"num_tokens": 6428579.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.625475597381592,
|
|
"epoch": 3.1377469300587295,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004428660372364959,
|
|
"loss": 6.2372,
|
|
"mean_token_accuracy": 0.14359409660100936,
|
|
"num_tokens": 6438976.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.640156650543213,
|
|
"epoch": 3.1430859583555795,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004425842380900532,
|
|
"loss": 6.1401,
|
|
"mean_token_accuracy": 0.13635687604546548,
|
|
"num_tokens": 6449919.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.540213251113892,
|
|
"epoch": 3.1484249866524294,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044230184729991565,
|
|
"loss": 6.1379,
|
|
"mean_token_accuracy": 0.14237724021077156,
|
|
"num_tokens": 6460874.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.662953329086304,
|
|
"epoch": 3.1537640149492794,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004420188658630428,
|
|
"loss": 6.1318,
|
|
"mean_token_accuracy": 0.14336148053407669,
|
|
"num_tokens": 6471606.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.579173994064331,
|
|
"epoch": 3.1591030432461293,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044173529477847974,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.140760400891304,
|
|
"num_tokens": 6481917.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.60531063079834,
|
|
"epoch": 3.1644420715429793,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044145113504735295,
|
|
"loss": 6.152,
|
|
"mean_token_accuracy": 0.14812284484505653,
|
|
"num_tokens": 6493295.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.603976917266846,
|
|
"epoch": 3.1697810998398293,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004411663876728674,
|
|
"loss": 6.1612,
|
|
"mean_token_accuracy": 0.1435246966779232,
|
|
"num_tokens": 6504208.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.5477807998657225,
|
|
"epoch": 3.175120128136679,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00044088105366030234,
|
|
"loss": 6.0951,
|
|
"mean_token_accuracy": 0.1487772673368454,
|
|
"num_tokens": 6515347.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.601194953918457,
|
|
"epoch": 3.180459156433529,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004405951340170084,
|
|
"loss": 6.1623,
|
|
"mean_token_accuracy": 0.13532078862190247,
|
|
"num_tokens": 6527674.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.564567041397095,
|
|
"epoch": 3.185798184730379,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00044030862975240364,
|
|
"loss": 6.1279,
|
|
"mean_token_accuracy": 0.1448235183954239,
|
|
"num_tokens": 6539048.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.595535612106323,
|
|
"epoch": 3.191137213027229,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044002154187797003,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.1410246193408966,
|
|
"num_tokens": 6550152.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.677579689025879,
|
|
"epoch": 3.196476241324079,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043973387140725,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.14386973679065704,
|
|
"num_tokens": 6561189.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.722335386276245,
|
|
"epoch": 3.201815269620929,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004394456193558428,
|
|
"loss": 6.2154,
|
|
"mean_token_accuracy": 0.13892951905727385,
|
|
"num_tokens": 6572548.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.201815269620929,
|
|
"eval_entropy": 6.4393722497023544,
|
|
"eval_loss": 6.7406086921691895,
|
|
"eval_mean_token_accuracy": 0.12297290774521889,
|
|
"eval_num_tokens": 6572548.0,
|
|
"eval_runtime": 1.8114,
|
|
"eval_samples_per_second": 1700.306,
|
|
"eval_steps_per_second": 212.538,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.626015567779541,
|
|
"epoch": 3.207154297917779,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004391567867414007,
|
|
"loss": 6.2295,
|
|
"mean_token_accuracy": 0.13846756219863893,
|
|
"num_tokens": 6584176.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.582659435272217,
|
|
"epoch": 3.212493326214629,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043886737458362596,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.15341638177633285,
|
|
"num_tokens": 6595070.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.523849010467529,
|
|
"epoch": 3.217832354511479,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004385773839042666,
|
|
"loss": 6.1148,
|
|
"mean_token_accuracy": 0.14375541731715202,
|
|
"num_tokens": 6606556.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.617887306213379,
|
|
"epoch": 3.223171382808329,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043828681572711316,
|
|
"loss": 6.0635,
|
|
"mean_token_accuracy": 0.14926337003707885,
|
|
"num_tokens": 6616944.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.489473152160644,
|
|
"epoch": 3.228510411105179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043799567107799504,
|
|
"loss": 6.1453,
|
|
"mean_token_accuracy": 0.15048904716968536,
|
|
"num_tokens": 6627541.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.605879449844361,
|
|
"epoch": 3.2338494394020287,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00043770395098477666,
|
|
"loss": 6.1314,
|
|
"mean_token_accuracy": 0.14454084560275077,
|
|
"num_tokens": 6639102.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.652470588684082,
|
|
"epoch": 3.2391884676988787,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004374116564773543,
|
|
"loss": 6.1631,
|
|
"mean_token_accuracy": 0.14270546436309814,
|
|
"num_tokens": 6650737.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.537238025665284,
|
|
"epoch": 3.2445274959957286,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004371187885876519,
|
|
"loss": 6.1842,
|
|
"mean_token_accuracy": 0.138457190990448,
|
|
"num_tokens": 6661812.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.6336798667907715,
|
|
"epoch": 3.2498665242925786,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043682534834961766,
|
|
"loss": 6.2576,
|
|
"mean_token_accuracy": 0.1425720900297165,
|
|
"num_tokens": 6672122.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.620451641082764,
|
|
"epoch": 3.2552055525894286,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043653133679922077,
|
|
"loss": 6.2141,
|
|
"mean_token_accuracy": 0.14090549126267432,
|
|
"num_tokens": 6684219.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.561987209320068,
|
|
"epoch": 3.2605445808862785,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043623675497444695,
|
|
"loss": 6.0454,
|
|
"mean_token_accuracy": 0.14704931527376175,
|
|
"num_tokens": 6694290.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.5945192813873295,
|
|
"epoch": 3.2658836091831285,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043594160391529545,
|
|
"loss": 6.157,
|
|
"mean_token_accuracy": 0.14867981225252153,
|
|
"num_tokens": 6704313.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.563622903823853,
|
|
"epoch": 3.271222637479979,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00043564588466377536,
|
|
"loss": 6.1617,
|
|
"mean_token_accuracy": 0.14300615936517716,
|
|
"num_tokens": 6715424.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.504296970367432,
|
|
"epoch": 3.2765616657768284,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004353495982639012,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.144931760430336,
|
|
"num_tokens": 6727025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.58922119140625,
|
|
"epoch": 3.281900694073679,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043505274576169037,
|
|
"loss": 6.1521,
|
|
"mean_token_accuracy": 0.14766536206007003,
|
|
"num_tokens": 6738344.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.5671083450317385,
|
|
"epoch": 3.2872397223705287,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00043475532820515836,
|
|
"loss": 6.1378,
|
|
"mean_token_accuracy": 0.14321689307689667,
|
|
"num_tokens": 6748812.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.579545545578003,
|
|
"epoch": 3.2925787506673787,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043445734664431575,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.14259575307369232,
|
|
"num_tokens": 6759776.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.557349061965942,
|
|
"epoch": 3.2979177789642287,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004341588021311644,
|
|
"loss": 6.1432,
|
|
"mean_token_accuracy": 0.14480729252099991,
|
|
"num_tokens": 6770702.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.5284239768981935,
|
|
"epoch": 3.3032568072610786,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043385969571969353,
|
|
"loss": 6.14,
|
|
"mean_token_accuracy": 0.1491951197385788,
|
|
"num_tokens": 6781599.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.5264880657196045,
|
|
"epoch": 3.3085958355579286,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043356002846587597,
|
|
"loss": 6.1786,
|
|
"mean_token_accuracy": 0.14892054349184036,
|
|
"num_tokens": 6792385.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.645450162887573,
|
|
"epoch": 3.3139348638547785,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004332598014276647,
|
|
"loss": 6.1443,
|
|
"mean_token_accuracy": 0.1407065249979496,
|
|
"num_tokens": 6803805.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.503142929077148,
|
|
"epoch": 3.3192738921516285,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00043295901566498907,
|
|
"loss": 6.1686,
|
|
"mean_token_accuracy": 0.1454508811235428,
|
|
"num_tokens": 6815460.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.593047142028809,
|
|
"epoch": 3.3246129204484784,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043265767223975076,
|
|
"loss": 6.1151,
|
|
"mean_token_accuracy": 0.1487767554819584,
|
|
"num_tokens": 6825995.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.522263669967652,
|
|
"epoch": 3.3299519487453284,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004323557722158204,
|
|
"loss": 6.0856,
|
|
"mean_token_accuracy": 0.14423727318644525,
|
|
"num_tokens": 6836793.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.56120867729187,
|
|
"epoch": 3.3352909770421784,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004320533166590335,
|
|
"loss": 6.1374,
|
|
"mean_token_accuracy": 0.14478903859853745,
|
|
"num_tokens": 6849063.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.615983200073242,
|
|
"epoch": 3.3406300053390283,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000431750306637187,
|
|
"loss": 6.1393,
|
|
"mean_token_accuracy": 0.13982861638069152,
|
|
"num_tokens": 6859821.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.547687005996704,
|
|
"epoch": 3.3459690336358783,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043144674322003525,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.14122217744588852,
|
|
"num_tokens": 6871026.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.6049974918365475,
|
|
"epoch": 3.3513080619327282,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004311426274792863,
|
|
"loss": 6.2092,
|
|
"mean_token_accuracy": 0.14053697511553764,
|
|
"num_tokens": 6883135.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.531812572479248,
|
|
"epoch": 3.356647090229578,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004308379604885981,
|
|
"loss": 6.1506,
|
|
"mean_token_accuracy": 0.14196842312812805,
|
|
"num_tokens": 6893943.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.480563163757324,
|
|
"epoch": 3.361986118526428,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004305327433235749,
|
|
"loss": 6.17,
|
|
"mean_token_accuracy": 0.14052093848586084,
|
|
"num_tokens": 6904172.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.660596752166748,
|
|
"epoch": 3.367325146823278,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004302269770617632,
|
|
"loss": 6.174,
|
|
"mean_token_accuracy": 0.14555246010422707,
|
|
"num_tokens": 6915440.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.630261611938477,
|
|
"epoch": 3.372664175120128,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000429920662782648,
|
|
"loss": 6.2339,
|
|
"mean_token_accuracy": 0.1399409957230091,
|
|
"num_tokens": 6927762.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.615788125991822,
|
|
"epoch": 3.378003203416978,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042961380156764925,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.1380746841430664,
|
|
"num_tokens": 6938139.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.513877248764038,
|
|
"epoch": 3.383342231713828,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004293063945001175,
|
|
"loss": 6.1746,
|
|
"mean_token_accuracy": 0.14551160410046576,
|
|
"num_tokens": 6949463.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.643848180770874,
|
|
"epoch": 3.388681260010678,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00042899844266533054,
|
|
"loss": 6.2151,
|
|
"mean_token_accuracy": 0.14019910618662834,
|
|
"num_tokens": 6960660.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.584480094909668,
|
|
"epoch": 3.394020288307528,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004286899471504896,
|
|
"loss": 6.1518,
|
|
"mean_token_accuracy": 0.14778395593166352,
|
|
"num_tokens": 6971283.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.641484785079956,
|
|
"epoch": 3.399359316604378,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042838090904471495,
|
|
"loss": 6.1865,
|
|
"mean_token_accuracy": 0.14859624207019806,
|
|
"num_tokens": 6981248.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.482734870910645,
|
|
"epoch": 3.404698344901228,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004280713294390428,
|
|
"loss": 6.153,
|
|
"mean_token_accuracy": 0.1452343285083771,
|
|
"num_tokens": 6991574.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.56207766532898,
|
|
"epoch": 3.4100373731980778,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004277612094264209,
|
|
"loss": 6.1664,
|
|
"mean_token_accuracy": 0.1509738191962242,
|
|
"num_tokens": 7002548.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.571753215789795,
|
|
"epoch": 3.415376401494928,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004274505501017048,
|
|
"loss": 6.1943,
|
|
"mean_token_accuracy": 0.14039533361792564,
|
|
"num_tokens": 7013815.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.619806480407715,
|
|
"epoch": 3.4207154297917777,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 6.1581,
|
|
"mean_token_accuracy": 0.145014002174139,
|
|
"num_tokens": 7024212.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.439695072174072,
|
|
"epoch": 3.426054458088628,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004268276179049292,
|
|
"loss": 6.1545,
|
|
"mean_token_accuracy": 0.14754147827625275,
|
|
"num_tokens": 7035167.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.619048595428467,
|
|
"epoch": 3.431393486385478,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004265153472320854,
|
|
"loss": 6.1243,
|
|
"mean_token_accuracy": 0.14939205199480057,
|
|
"num_tokens": 7045036.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.576473379135132,
|
|
"epoch": 3.436732514682328,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042620254164557137,
|
|
"loss": 6.1894,
|
|
"mean_token_accuracy": 0.14514195621013642,
|
|
"num_tokens": 7054938.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.575670194625855,
|
|
"epoch": 3.442071542979178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000425889202249724,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.14288137555122377,
|
|
"num_tokens": 7065645.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.573948764801026,
|
|
"epoch": 3.447410571276028,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042557533015076456,
|
|
"loss": 6.1069,
|
|
"mean_token_accuracy": 0.14459412023425103,
|
|
"num_tokens": 7075909.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.556999588012696,
|
|
"epoch": 3.452749599572878,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042526092645679535,
|
|
"loss": 6.1364,
|
|
"mean_token_accuracy": 0.15125184655189514,
|
|
"num_tokens": 7085761.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.537608480453491,
|
|
"epoch": 3.458088627869728,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042494599227779523,
|
|
"loss": 6.2128,
|
|
"mean_token_accuracy": 0.1416289672255516,
|
|
"num_tokens": 7096357.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.5380041122436525,
|
|
"epoch": 3.4634276561665778,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042463052872561587,
|
|
"loss": 6.2078,
|
|
"mean_token_accuracy": 0.14668214172124863,
|
|
"num_tokens": 7107534.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.580185413360596,
|
|
"epoch": 3.4687666844634277,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042431453691397787,
|
|
"loss": 6.2045,
|
|
"mean_token_accuracy": 0.13919102773070335,
|
|
"num_tokens": 7119186.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.587459516525269,
|
|
"epoch": 3.4741057127602777,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004239980179584669,
|
|
"loss": 6.0809,
|
|
"mean_token_accuracy": 0.1546573482453823,
|
|
"num_tokens": 7130187.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.536509609222412,
|
|
"epoch": 3.4794447410571276,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004236809729765295,
|
|
"loss": 6.1595,
|
|
"mean_token_accuracy": 0.14584122598171234,
|
|
"num_tokens": 7140734.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.479880380630493,
|
|
"epoch": 3.4847837693539776,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042336340308746944,
|
|
"loss": 6.1511,
|
|
"mean_token_accuracy": 0.1437065228819847,
|
|
"num_tokens": 7151642.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.549475431442261,
|
|
"epoch": 3.4901227976508276,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004230453094124436,
|
|
"loss": 6.1861,
|
|
"mean_token_accuracy": 0.13861185908317566,
|
|
"num_tokens": 7162650.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.549925708770752,
|
|
"epoch": 3.4954618259476775,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042272669307445823,
|
|
"loss": 6.161,
|
|
"mean_token_accuracy": 0.14364441037178038,
|
|
"num_tokens": 7174095.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.569450330734253,
|
|
"epoch": 3.5008008542445275,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00042240755519836436,
|
|
"loss": 6.1446,
|
|
"mean_token_accuracy": 0.15000987201929092,
|
|
"num_tokens": 7185923.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.502327013015747,
|
|
"epoch": 3.5061398825413774,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042208789691085464,
|
|
"loss": 6.1405,
|
|
"mean_token_accuracy": 0.14843530356884002,
|
|
"num_tokens": 7197287.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.547327613830566,
|
|
"epoch": 3.5114789108382274,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000421767719340459,
|
|
"loss": 6.1417,
|
|
"mean_token_accuracy": 0.14269062951207162,
|
|
"num_tokens": 7208143.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.547702884674072,
|
|
"epoch": 3.5168179391350773,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004214470236175404,
|
|
"loss": 6.1023,
|
|
"mean_token_accuracy": 0.15332724750041962,
|
|
"num_tokens": 7218516.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.547658395767212,
|
|
"epoch": 3.5221569674319273,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042112581087429133,
|
|
"loss": 6.1804,
|
|
"mean_token_accuracy": 0.14948860853910445,
|
|
"num_tokens": 7229066.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.561883544921875,
|
|
"epoch": 3.5274959957287773,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00042080408224472946,
|
|
"loss": 6.1691,
|
|
"mean_token_accuracy": 0.14923621192574502,
|
|
"num_tokens": 7239612.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.505513095855713,
|
|
"epoch": 3.532835024025627,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042048183886469386,
|
|
"loss": 6.08,
|
|
"mean_token_accuracy": 0.14673556685447692,
|
|
"num_tokens": 7250266.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.513124561309814,
|
|
"epoch": 3.538174052322477,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042015908187184077,
|
|
"loss": 6.1603,
|
|
"mean_token_accuracy": 0.14263225272297858,
|
|
"num_tokens": 7261183.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.538923549652099,
|
|
"epoch": 3.543513080619327,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041983581240563985,
|
|
"loss": 6.1348,
|
|
"mean_token_accuracy": 0.1486171156167984,
|
|
"num_tokens": 7271557.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.535774898529053,
|
|
"epoch": 3.5488521089161775,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004195120316073698,
|
|
"loss": 6.1998,
|
|
"mean_token_accuracy": 0.1453086718916893,
|
|
"num_tokens": 7282540.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.511842823028564,
|
|
"epoch": 3.554191137213027,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041918774062011467,
|
|
"loss": 6.1107,
|
|
"mean_token_accuracy": 0.14593104943633078,
|
|
"num_tokens": 7293568.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.523481845855713,
|
|
"epoch": 3.5595301655098774,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00041886294058875975,
|
|
"loss": 6.2032,
|
|
"mean_token_accuracy": 0.1386525124311447,
|
|
"num_tokens": 7304745.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.553685283660888,
|
|
"epoch": 3.564869193806727,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004185376326599874,
|
|
"loss": 6.2036,
|
|
"mean_token_accuracy": 0.14021066576242447,
|
|
"num_tokens": 7316547.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.578783273696899,
|
|
"epoch": 3.5702082221035774,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041821181798227305,
|
|
"loss": 6.2193,
|
|
"mean_token_accuracy": 0.14152044877409936,
|
|
"num_tokens": 7328493.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.494209146499633,
|
|
"epoch": 3.575547250400427,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004178854977058812,
|
|
"loss": 6.1514,
|
|
"mean_token_accuracy": 0.14675125181674958,
|
|
"num_tokens": 7338831.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.58004937171936,
|
|
"epoch": 3.5808862786972773,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004175586729828614,
|
|
"loss": 6.2568,
|
|
"mean_token_accuracy": 0.13772836774587632,
|
|
"num_tokens": 7350884.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.56378321647644,
|
|
"epoch": 3.5862253069941272,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004172313449670439,
|
|
"loss": 6.1245,
|
|
"mean_token_accuracy": 0.14097074419260025,
|
|
"num_tokens": 7360674.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.479941940307617,
|
|
"epoch": 3.591564335290977,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00041690351481403606,
|
|
"loss": 6.1567,
|
|
"mean_token_accuracy": 0.1383296102285385,
|
|
"num_tokens": 7371965.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.642097520828247,
|
|
"epoch": 3.596903363587827,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004165751836812178,
|
|
"loss": 6.241,
|
|
"mean_token_accuracy": 0.145929317176342,
|
|
"num_tokens": 7383649.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.6187879085540775,
|
|
"epoch": 3.602242391884677,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004162463527277378,
|
|
"loss": 6.1849,
|
|
"mean_token_accuracy": 0.1447222724556923,
|
|
"num_tokens": 7393692.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.562582874298096,
|
|
"epoch": 3.607581420181527,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00041591702311450924,
|
|
"loss": 6.2116,
|
|
"mean_token_accuracy": 0.13732682764530182,
|
|
"num_tokens": 7404149.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.533840751647949,
|
|
"epoch": 3.612920448478377,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004155871960042059,
|
|
"loss": 6.1192,
|
|
"mean_token_accuracy": 0.1485688403248787,
|
|
"num_tokens": 7415011.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.53869137763977,
|
|
"epoch": 3.618259476775227,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00041525687256125786,
|
|
"loss": 6.1796,
|
|
"mean_token_accuracy": 0.14087817072868347,
|
|
"num_tokens": 7426297.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.520381307601928,
|
|
"epoch": 3.623598505072077,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004149260539518475,
|
|
"loss": 6.054,
|
|
"mean_token_accuracy": 0.1445208042860031,
|
|
"num_tokens": 7437718.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.488810968399048,
|
|
"epoch": 3.628937533368927,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041459474134390527,
|
|
"loss": 6.1478,
|
|
"mean_token_accuracy": 0.1479748159646988,
|
|
"num_tokens": 7448395.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.43403959274292,
|
|
"epoch": 3.634276561665777,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041426293590710587,
|
|
"loss": 6.1326,
|
|
"mean_token_accuracy": 0.15362447500228882,
|
|
"num_tokens": 7458374.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.588842344284058,
|
|
"epoch": 3.639615589962627,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00041393063881286346,
|
|
"loss": 6.1325,
|
|
"mean_token_accuracy": 0.15165682807564734,
|
|
"num_tokens": 7469386.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.571090030670166,
|
|
"epoch": 3.6449546182594768,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004135978512343284,
|
|
"loss": 6.1911,
|
|
"mean_token_accuracy": 0.1359236106276512,
|
|
"num_tokens": 7479788.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.470835828781128,
|
|
"epoch": 3.6502936465563267,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00041326457434638246,
|
|
"loss": 6.1273,
|
|
"mean_token_accuracy": 0.14886193200945855,
|
|
"num_tokens": 7490329.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.532852983474731,
|
|
"epoch": 3.6556326748531767,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000412930809325635,
|
|
"loss": 6.175,
|
|
"mean_token_accuracy": 0.139870285987854,
|
|
"num_tokens": 7502206.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.530126094818115,
|
|
"epoch": 3.6609717031500266,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00041259655735041847,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.14210368692874908,
|
|
"num_tokens": 7513704.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.4876954555511475,
|
|
"epoch": 3.6663107314468766,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00041226181960078473,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.1456264317035675,
|
|
"num_tokens": 7524320.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.542640733718872,
|
|
"epoch": 3.6716497597437265,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004119265972585005,
|
|
"loss": 6.1752,
|
|
"mean_token_accuracy": 0.14658585488796233,
|
|
"num_tokens": 7533962.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.579928541183472,
|
|
"epoch": 3.6769887880405765,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00041159089150704325,
|
|
"loss": 6.2128,
|
|
"mean_token_accuracy": 0.14553995281457902,
|
|
"num_tokens": 7545841.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.596597528457641,
|
|
"epoch": 3.6823278163374265,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041125470353159735,
|
|
"loss": 6.1607,
|
|
"mean_token_accuracy": 0.14802061766386032,
|
|
"num_tokens": 7555533.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.503862905502319,
|
|
"epoch": 3.6876668446342764,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00041091803451904927,
|
|
"loss": 6.1554,
|
|
"mean_token_accuracy": 0.1387634038925171,
|
|
"num_tokens": 7566228.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.505353736877441,
|
|
"epoch": 3.693005872931127,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00041058088565798403,
|
|
"loss": 6.198,
|
|
"mean_token_accuracy": 0.14261291325092315,
|
|
"num_tokens": 7577694.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.62959246635437,
|
|
"epoch": 3.6983449012279763,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041024325813868065,
|
|
"loss": 6.1753,
|
|
"mean_token_accuracy": 0.14205058813095092,
|
|
"num_tokens": 7588141.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.598307275772095,
|
|
"epoch": 3.7036839295248267,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00040990515315310794,
|
|
"loss": 6.1758,
|
|
"mean_token_accuracy": 0.14348702132701874,
|
|
"num_tokens": 7599831.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.537150287628174,
|
|
"epoch": 3.7090229578216762,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040956657189492046,
|
|
"loss": 6.0966,
|
|
"mean_token_accuracy": 0.14398248717188836,
|
|
"num_tokens": 7609567.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.556712341308594,
|
|
"epoch": 3.7143619861185266,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004092275155594541,
|
|
"loss": 6.2319,
|
|
"mean_token_accuracy": 0.13728955760598183,
|
|
"num_tokens": 7620722.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.567393589019775,
|
|
"epoch": 3.719701014415376,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004088879853437221,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.13819843381643296,
|
|
"num_tokens": 7633036.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.479170513153076,
|
|
"epoch": 3.7250400427122266,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040854798244641067,
|
|
"loss": 6.1627,
|
|
"mean_token_accuracy": 0.14585598707199096,
|
|
"num_tokens": 7642789.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.506835269927978,
|
|
"epoch": 3.7303790710090765,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004082075080678748,
|
|
"loss": 6.1908,
|
|
"mean_token_accuracy": 0.14425090700387955,
|
|
"num_tokens": 7654234.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.539835405349732,
|
|
"epoch": 3.7357180993059265,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040786656341013376,
|
|
"loss": 6.1273,
|
|
"mean_token_accuracy": 0.14244442880153657,
|
|
"num_tokens": 7665994.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7357180993059265,
|
|
"eval_entropy": 6.402802858724223,
|
|
"eval_loss": 6.652650833129883,
|
|
"eval_mean_token_accuracy": 0.12644047731315936,
|
|
"eval_num_tokens": 7665994.0,
|
|
"eval_runtime": 1.8024,
|
|
"eval_samples_per_second": 1708.804,
|
|
"eval_steps_per_second": 213.6,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.511250829696655,
|
|
"epoch": 3.7410571276027764,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040752514967686747,
|
|
"loss": 6.084,
|
|
"mean_token_accuracy": 0.14790418446063996,
|
|
"num_tokens": 7676813.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.438376808166504,
|
|
"epoch": 3.7463961558996264,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004071832680734118,
|
|
"loss": 6.1267,
|
|
"mean_token_accuracy": 0.1470203071832657,
|
|
"num_tokens": 7687775.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.558165264129639,
|
|
"epoch": 3.7517351841964763,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040684091980675426,
|
|
"loss": 6.2229,
|
|
"mean_token_accuracy": 0.1389820992946625,
|
|
"num_tokens": 7698909.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.513540887832642,
|
|
"epoch": 3.7570742124933263,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040649810608552994,
|
|
"loss": 6.132,
|
|
"mean_token_accuracy": 0.1510733187198639,
|
|
"num_tokens": 7708783.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.544998741149902,
|
|
"epoch": 3.7624132407901762,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040615482812001726,
|
|
"loss": 6.174,
|
|
"mean_token_accuracy": 0.14355052411556243,
|
|
"num_tokens": 7719010.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.5300548553466795,
|
|
"epoch": 3.767752269087026,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004058110871221335,
|
|
"loss": 6.2019,
|
|
"mean_token_accuracy": 0.1421898864209652,
|
|
"num_tokens": 7730231.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.484360885620117,
|
|
"epoch": 3.773091297383876,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040546688430543074,
|
|
"loss": 6.1209,
|
|
"mean_token_accuracy": 0.14670493602752685,
|
|
"num_tokens": 7741656.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.542676115036011,
|
|
"epoch": 3.778430325680726,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00040512222088509147,
|
|
"loss": 6.1385,
|
|
"mean_token_accuracy": 0.1478129133582115,
|
|
"num_tokens": 7751956.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.508578777313232,
|
|
"epoch": 3.783769353977576,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004047770980779242,
|
|
"loss": 6.1029,
|
|
"mean_token_accuracy": 0.1496267482638359,
|
|
"num_tokens": 7761944.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.446523857116699,
|
|
"epoch": 3.789108382274426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004044315171023593,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.14807819575071335,
|
|
"num_tokens": 7771370.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.563878965377808,
|
|
"epoch": 3.794447410571276,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004040854791784448,
|
|
"loss": 6.2957,
|
|
"mean_token_accuracy": 0.13254996612668038,
|
|
"num_tokens": 7783615.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.600985717773438,
|
|
"epoch": 3.799786438868126,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004037389855278419,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.1492430627346039,
|
|
"num_tokens": 7794236.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.468111991882324,
|
|
"epoch": 3.805125467164976,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040339203737382065,
|
|
"loss": 6.166,
|
|
"mean_token_accuracy": 0.14735100716352462,
|
|
"num_tokens": 7806158.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.527578115463257,
|
|
"epoch": 3.810464495461826,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004030446359412556,
|
|
"loss": 6.12,
|
|
"mean_token_accuracy": 0.14553360044956207,
|
|
"num_tokens": 7818283.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.579574012756348,
|
|
"epoch": 3.815803523758676,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004026967824566218,
|
|
"loss": 6.1631,
|
|
"mean_token_accuracy": 0.1417527861893177,
|
|
"num_tokens": 7828683.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.49411449432373,
|
|
"epoch": 3.8211425520555258,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004023484781479902,
|
|
"loss": 6.1374,
|
|
"mean_token_accuracy": 0.14596590846776963,
|
|
"num_tokens": 7839116.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.544578886032104,
|
|
"epoch": 3.8264815803523757,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.1430832676589489,
|
|
"num_tokens": 7850300.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.535199403762817,
|
|
"epoch": 3.8318206086492257,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004016505219789703,
|
|
"loss": 6.096,
|
|
"mean_token_accuracy": 0.145581466704607,
|
|
"num_tokens": 7861432.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.481894445419312,
|
|
"epoch": 3.837159636946076,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004013008725826644,
|
|
"loss": 6.19,
|
|
"mean_token_accuracy": 0.145276015996933,
|
|
"num_tokens": 7872472.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.576520872116089,
|
|
"epoch": 3.8424986652429256,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004009507772905167,
|
|
"loss": 6.1948,
|
|
"mean_token_accuracy": 0.146095909178257,
|
|
"num_tokens": 7883554.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.5780809879302975,
|
|
"epoch": 3.847837693539776,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00040060023733851254,
|
|
"loss": 6.1698,
|
|
"mean_token_accuracy": 0.14025285094976425,
|
|
"num_tokens": 7894853.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.464986371994018,
|
|
"epoch": 3.8531767218366255,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040024925396420707,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.1404825381934643,
|
|
"num_tokens": 7905459.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.486498022079468,
|
|
"epoch": 3.858515750133476,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00039989782840672123,
|
|
"loss": 6.0861,
|
|
"mean_token_accuracy": 0.14591265097260475,
|
|
"num_tokens": 7916646.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.47977614402771,
|
|
"epoch": 3.8638547784303254,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003995459619067366,
|
|
"loss": 6.1397,
|
|
"mean_token_accuracy": 0.14808139503002166,
|
|
"num_tokens": 7928119.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.526146650314331,
|
|
"epoch": 3.869193806727176,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039919365570649187,
|
|
"loss": 6.2565,
|
|
"mean_token_accuracy": 0.13531762585043908,
|
|
"num_tokens": 7939295.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.55569486618042,
|
|
"epoch": 3.874532835024026,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00039884091104977774,
|
|
"loss": 6.0711,
|
|
"mean_token_accuracy": 0.14545215293765068,
|
|
"num_tokens": 7950094.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.515868520736694,
|
|
"epoch": 3.8798718633208757,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039848772918193315,
|
|
"loss": 6.0973,
|
|
"mean_token_accuracy": 0.1533176302909851,
|
|
"num_tokens": 7960767.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.44457221031189,
|
|
"epoch": 3.8852108916177257,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003981341113498404,
|
|
"loss": 6.1581,
|
|
"mean_token_accuracy": 0.14428275674581528,
|
|
"num_tokens": 7971660.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.5656482696533205,
|
|
"epoch": 3.8905499199145757,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003977800588019209,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.1490553878247738,
|
|
"num_tokens": 7983337.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.527936792373657,
|
|
"epoch": 3.8958889482114256,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0003974255727881307,
|
|
"loss": 6.1845,
|
|
"mean_token_accuracy": 0.14226604625582695,
|
|
"num_tokens": 7994540.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.515545797348023,
|
|
"epoch": 3.9012279765082756,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003970706545599566,
|
|
"loss": 6.0988,
|
|
"mean_token_accuracy": 0.1402305468916893,
|
|
"num_tokens": 8005396.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.4922425746917725,
|
|
"epoch": 3.9065670048051255,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039671530537041094,
|
|
"loss": 6.1029,
|
|
"mean_token_accuracy": 0.15106893181800843,
|
|
"num_tokens": 8015411.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.487470006942749,
|
|
"epoch": 3.9119060331019755,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039635952647402756,
|
|
"loss": 6.0656,
|
|
"mean_token_accuracy": 0.15119688287377359,
|
|
"num_tokens": 8025720.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.427022790908813,
|
|
"epoch": 3.9172450613988254,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003960033191268574,
|
|
"loss": 6.1439,
|
|
"mean_token_accuracy": 0.1427379496395588,
|
|
"num_tokens": 8036081.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.473222827911377,
|
|
"epoch": 3.9225840896956754,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003956466845864641,
|
|
"loss": 6.0318,
|
|
"mean_token_accuracy": 0.14493608772754668,
|
|
"num_tokens": 8046714.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.415006065368653,
|
|
"epoch": 3.9279231179925254,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003952896241119194,
|
|
"loss": 6.0935,
|
|
"mean_token_accuracy": 0.15388759225606918,
|
|
"num_tokens": 8057174.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.449224328994751,
|
|
"epoch": 3.9332621462893753,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003949321389637986,
|
|
"loss": 6.0499,
|
|
"mean_token_accuracy": 0.15181737542152404,
|
|
"num_tokens": 8067697.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.493015480041504,
|
|
"epoch": 3.9386011745862253,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003945742304041766,
|
|
"loss": 6.1459,
|
|
"mean_token_accuracy": 0.14583780020475387,
|
|
"num_tokens": 8077800.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.556552982330322,
|
|
"epoch": 3.9439402028830752,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0003942158996966229,
|
|
"loss": 6.2258,
|
|
"mean_token_accuracy": 0.1387764908373356,
|
|
"num_tokens": 8090319.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.55411057472229,
|
|
"epoch": 3.949279231179925,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003938571481061975,
|
|
"loss": 6.1572,
|
|
"mean_token_accuracy": 0.14131653159856797,
|
|
"num_tokens": 8100498.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.481984329223633,
|
|
"epoch": 3.954618259476775,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039349797689944613,
|
|
"loss": 6.0891,
|
|
"mean_token_accuracy": 0.14142293035984038,
|
|
"num_tokens": 8110506.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.5264497756958,
|
|
"epoch": 3.959957287773625,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00039313838734439605,
|
|
"loss": 6.1093,
|
|
"mean_token_accuracy": 0.14747968167066575,
|
|
"num_tokens": 8121999.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.481518793106079,
|
|
"epoch": 3.965296316070475,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00039277838071055145,
|
|
"loss": 6.1913,
|
|
"mean_token_accuracy": 0.14093517512083054,
|
|
"num_tokens": 8132884.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.576885175704956,
|
|
"epoch": 3.970635344367325,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000392417958268889,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.14702757447957993,
|
|
"num_tokens": 8143344.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.490785789489746,
|
|
"epoch": 3.975974372664175,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039205712129185337,
|
|
"loss": 6.1281,
|
|
"mean_token_accuracy": 0.14834580570459366,
|
|
"num_tokens": 8154658.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.4275837421417235,
|
|
"epoch": 3.981313400961025,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003916958710533528,
|
|
"loss": 6.082,
|
|
"mean_token_accuracy": 0.14838915020227433,
|
|
"num_tokens": 8166057.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.574824666976928,
|
|
"epoch": 3.986652429257875,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003913342088287541,
|
|
"loss": 6.106,
|
|
"mean_token_accuracy": 0.14580099284648895,
|
|
"num_tokens": 8176470.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.523640108108521,
|
|
"epoch": 3.9919914575547253,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000390972135894879,
|
|
"loss": 6.1629,
|
|
"mean_token_accuracy": 0.13956414237618447,
|
|
"num_tokens": 8186458.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.550470685958862,
|
|
"epoch": 3.997330485851575,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039060965352999926,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.13434221893548964,
|
|
"num_tokens": 8197826.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.523906389872233,
|
|
"epoch": 4.00213561131874,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003902467630138317,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.14978978451755312,
|
|
"num_tokens": 8208809.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.455614471435547,
|
|
"epoch": 4.00747463961559,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038988346562753435,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.16545109897851945,
|
|
"num_tokens": 8218740.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.479466819763184,
|
|
"epoch": 4.0128136679124395,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00038951976265370176,
|
|
"loss": 5.8186,
|
|
"mean_token_accuracy": 0.1650289312005043,
|
|
"num_tokens": 8230360.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.4658434867858885,
|
|
"epoch": 4.01815269620929,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00038915565537636003,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.16136547774076462,
|
|
"num_tokens": 8240886.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.439322376251221,
|
|
"epoch": 4.0234917245061395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000388791145080963,
|
|
"loss": 5.8101,
|
|
"mean_token_accuracy": 0.16565286070108415,
|
|
"num_tokens": 8252103.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.506541919708252,
|
|
"epoch": 4.02883075280299,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003884262330543872,
|
|
"loss": 5.9089,
|
|
"mean_token_accuracy": 0.15189366191625595,
|
|
"num_tokens": 8262890.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.481657361984253,
|
|
"epoch": 4.03416978109984,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00038806092058492727,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.15886562168598176,
|
|
"num_tokens": 8273562.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.5099996566772464,
|
|
"epoch": 4.03950880939669,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003876952089622918,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.15011855959892273,
|
|
"num_tokens": 8285104.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.495954179763794,
|
|
"epoch": 4.04484783769354,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003873290994775984,
|
|
"loss": 5.8799,
|
|
"mean_token_accuracy": 0.15607954263687135,
|
|
"num_tokens": 8296760.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.490471601486206,
|
|
"epoch": 4.05018686599039,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00038696259342336966,
|
|
"loss": 5.8882,
|
|
"mean_token_accuracy": 0.16234036087989806,
|
|
"num_tokens": 8306929.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.442892169952392,
|
|
"epoch": 4.05552589428724,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038659569209352766,
|
|
"loss": 5.9147,
|
|
"mean_token_accuracy": 0.15829430669546127,
|
|
"num_tokens": 8316736.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.505135583877563,
|
|
"epoch": 4.06086492258409,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00038622839678339034,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.157439686357975,
|
|
"num_tokens": 8327688.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.4566991329193115,
|
|
"epoch": 4.06620395088094,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038586070878966663,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.16600643396377562,
|
|
"num_tokens": 8337986.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.422807598114014,
|
|
"epoch": 4.0715429791777895,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003854926294104514,
|
|
"loss": 5.8251,
|
|
"mean_token_accuracy": 0.1629556953907013,
|
|
"num_tokens": 8348473.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.492142152786255,
|
|
"epoch": 4.07688200747464,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003851241599452218,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.15974331200122832,
|
|
"num_tokens": 8359563.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.476008796691895,
|
|
"epoch": 4.082221035771489,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038475530169483185,
|
|
"loss": 5.941,
|
|
"mean_token_accuracy": 0.15269246846437454,
|
|
"num_tokens": 8370965.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.416006135940552,
|
|
"epoch": 4.08756006406834,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00038438605596150807,
|
|
"loss": 5.9162,
|
|
"mean_token_accuracy": 0.15478795990347863,
|
|
"num_tokens": 8382606.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.487174367904663,
|
|
"epoch": 4.092899092365189,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038401642404884513,
|
|
"loss": 5.8667,
|
|
"mean_token_accuracy": 0.1620238795876503,
|
|
"num_tokens": 8392618.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.451450061798096,
|
|
"epoch": 4.09823812066204,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038364640726180096,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.15819164365530014,
|
|
"num_tokens": 8403210.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.457551288604736,
|
|
"epoch": 4.103577148958889,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038327600690669244,
|
|
"loss": 5.9271,
|
|
"mean_token_accuracy": 0.15056874752044677,
|
|
"num_tokens": 8413838.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.452711200714111,
|
|
"epoch": 4.10891617725574,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00038290522429119034,
|
|
"loss": 5.8383,
|
|
"mean_token_accuracy": 0.1614674672484398,
|
|
"num_tokens": 8425282.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.410413837432861,
|
|
"epoch": 4.114255205552589,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038253406072431525,
|
|
"loss": 5.8392,
|
|
"mean_token_accuracy": 0.16616302728652954,
|
|
"num_tokens": 8436762.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.500837230682373,
|
|
"epoch": 4.11959423384944,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038216251751643245,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.15862398594617844,
|
|
"num_tokens": 8448636.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.445639896392822,
|
|
"epoch": 4.124933262146289,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003817905959792477,
|
|
"loss": 5.8186,
|
|
"mean_token_accuracy": 0.16568127423524856,
|
|
"num_tokens": 8459730.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.432496213912964,
|
|
"epoch": 4.1302722904431395,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003814182974258023,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.16377303451299668,
|
|
"num_tokens": 8470628.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.410172033309936,
|
|
"epoch": 4.135611318739989,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038104562317046873,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.1582864850759506,
|
|
"num_tokens": 8480785.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.400460815429687,
|
|
"epoch": 4.140950347036839,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003806725745289455,
|
|
"loss": 5.7772,
|
|
"mean_token_accuracy": 0.16919722110033036,
|
|
"num_tokens": 8491596.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.427718925476074,
|
|
"epoch": 4.146289375333689,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038029915281825334,
|
|
"loss": 5.8855,
|
|
"mean_token_accuracy": 0.1604493737220764,
|
|
"num_tokens": 8502458.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.4377576351165775,
|
|
"epoch": 4.151628403630539,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00037992535935672974,
|
|
"loss": 5.8356,
|
|
"mean_token_accuracy": 0.1650594249367714,
|
|
"num_tokens": 8513018.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.484112358093261,
|
|
"epoch": 4.156967431927389,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00037955119546402474,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.16013461947441102,
|
|
"num_tokens": 8523757.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.320549201965332,
|
|
"epoch": 4.162306460224239,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003791766624610961,
|
|
"loss": 5.7758,
|
|
"mean_token_accuracy": 0.16299208998680115,
|
|
"num_tokens": 8535446.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.370971965789795,
|
|
"epoch": 4.167645488521089,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003788017616702048,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.168772192299366,
|
|
"num_tokens": 8545868.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.428561162948609,
|
|
"epoch": 4.172984516817939,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003784264944149102,
|
|
"loss": 5.8338,
|
|
"mean_token_accuracy": 0.15697346329689027,
|
|
"num_tokens": 8556465.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.4201459884643555,
|
|
"epoch": 4.1783235451147895,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003780508620200655,
|
|
"loss": 5.866,
|
|
"mean_token_accuracy": 0.16035554856061934,
|
|
"num_tokens": 8568227.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.408050775527954,
|
|
"epoch": 4.183662573411639,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003776748658118128,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.16062171459198,
|
|
"num_tokens": 8579309.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.407215690612793,
|
|
"epoch": 4.1890016017084895,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00037729850711757903,
|
|
"loss": 5.8097,
|
|
"mean_token_accuracy": 0.16254528909921645,
|
|
"num_tokens": 8590487.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.501875162124634,
|
|
"epoch": 4.194340630005339,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003769217872660703,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.15888245403766632,
|
|
"num_tokens": 8600940.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.375162887573242,
|
|
"epoch": 4.199679658302189,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00037654470758726816,
|
|
"loss": 5.6835,
|
|
"mean_token_accuracy": 0.1780616506934166,
|
|
"num_tokens": 8611322.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.3994794368743895,
|
|
"epoch": 4.205018686599039,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037616726941242445,
|
|
"loss": 5.8733,
|
|
"mean_token_accuracy": 0.15676409006118774,
|
|
"num_tokens": 8622055.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.383853530883789,
|
|
"epoch": 4.210357714895889,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003757894740740565,
|
|
"loss": 5.9048,
|
|
"mean_token_accuracy": 0.16065036207437516,
|
|
"num_tokens": 8633515.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.382141780853272,
|
|
"epoch": 4.215696743192739,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003754113229059426,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.16290465593338013,
|
|
"num_tokens": 8644199.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.386103391647339,
|
|
"epoch": 4.221035771489589,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003750328172431176,
|
|
"loss": 5.902,
|
|
"mean_token_accuracy": 0.15742135792970657,
|
|
"num_tokens": 8655025.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.384013700485229,
|
|
"epoch": 4.226374799786439,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003746539584218673,
|
|
"loss": 5.852,
|
|
"mean_token_accuracy": 0.16765398383140565,
|
|
"num_tokens": 8666251.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.393583822250366,
|
|
"epoch": 4.231713828083289,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.1669989585876465,
|
|
"num_tokens": 8676727.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.466755676269531,
|
|
"epoch": 4.237052856380139,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000373895186655465,
|
|
"loss": 5.8869,
|
|
"mean_token_accuracy": 0.16504256725311278,
|
|
"num_tokens": 8687498.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.458343982696533,
|
|
"epoch": 4.242391884676989,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00037351527638910035,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.16048015058040618,
|
|
"num_tokens": 8698505.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.432484912872314,
|
|
"epoch": 4.2477309129738385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003731350183218755,
|
|
"loss": 5.9565,
|
|
"mean_token_accuracy": 0.1561967760324478,
|
|
"num_tokens": 8709349.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.471627855300904,
|
|
"epoch": 4.253069941270689,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003727544137962637,
|
|
"loss": 5.7887,
|
|
"mean_token_accuracy": 0.1591472879052162,
|
|
"num_tokens": 8719515.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.417425298690796,
|
|
"epoch": 4.2584089695675384,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003723734641559607,
|
|
"loss": 5.9763,
|
|
"mean_token_accuracy": 0.15346258357167245,
|
|
"num_tokens": 8731030.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.447788286209106,
|
|
"epoch": 4.263747997864389,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003719921707458809,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.15848195552825928,
|
|
"num_tokens": 8742776.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.39237151145935,
|
|
"epoch": 4.269087026161238,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003716105349121523,
|
|
"loss": 5.8143,
|
|
"mean_token_accuracy": 0.1576450362801552,
|
|
"num_tokens": 8753994.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.269087026161238,
|
|
"eval_entropy": 6.213875951395407,
|
|
"eval_loss": 6.626263618469238,
|
|
"eval_mean_token_accuracy": 0.12890922446529587,
|
|
"eval_num_tokens": 8753994.0,
|
|
"eval_runtime": 1.8138,
|
|
"eval_samples_per_second": 1698.115,
|
|
"eval_steps_per_second": 212.264,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9360,
|
|
"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": 1898878099230720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|