6c35b32e2a
Model: fpadovani/heb-hebr-10mb-after-ppt-Dp-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": 8.144560241699219,
|
|
"epoch": 0.005339028296849973,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.8897,
|
|
"mean_token_accuracy": 0.06864776574075222,
|
|
"num_tokens": 10658.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 8.15593843460083,
|
|
"epoch": 0.010678056593699947,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.8846,
|
|
"mean_token_accuracy": 0.06999277919530869,
|
|
"num_tokens": 21327.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 8.144119453430175,
|
|
"epoch": 0.01601708489054992,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.9107,
|
|
"mean_token_accuracy": 0.06488565169274807,
|
|
"num_tokens": 32403.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 8.17639331817627,
|
|
"epoch": 0.021356113187399894,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.8553,
|
|
"mean_token_accuracy": 0.0711637206375599,
|
|
"num_tokens": 43471.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.177908802032471,
|
|
"epoch": 0.026695141484249868,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.8379,
|
|
"mean_token_accuracy": 0.07274281270802022,
|
|
"num_tokens": 53724.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.16096534729004,
|
|
"epoch": 0.03203416978109984,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.9025,
|
|
"mean_token_accuracy": 0.0708409644663334,
|
|
"num_tokens": 65115.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 8.17019271850586,
|
|
"epoch": 0.03737319807794982,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.8468,
|
|
"mean_token_accuracy": 0.07302796579897404,
|
|
"num_tokens": 75902.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.207113361358642,
|
|
"epoch": 0.04271222637479979,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.8962,
|
|
"mean_token_accuracy": 0.06636322438716888,
|
|
"num_tokens": 87051.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 8.1253080368042,
|
|
"epoch": 0.04805125467164976,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.8199,
|
|
"mean_token_accuracy": 0.07184852361679077,
|
|
"num_tokens": 98998.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 8.162546825408935,
|
|
"epoch": 0.053390282968499736,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.8969,
|
|
"mean_token_accuracy": 0.07030453905463219,
|
|
"num_tokens": 111207.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 8.178112983703613,
|
|
"epoch": 0.05872931126534971,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.8128,
|
|
"mean_token_accuracy": 0.07619077749550343,
|
|
"num_tokens": 121039.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 8.193019104003906,
|
|
"epoch": 0.06406833956219968,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.9013,
|
|
"mean_token_accuracy": 0.07129124961793423,
|
|
"num_tokens": 132604.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 8.207361793518066,
|
|
"epoch": 0.06940736785904965,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.8666,
|
|
"mean_token_accuracy": 0.07484896555542946,
|
|
"num_tokens": 142825.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 8.12415533065796,
|
|
"epoch": 0.07474639615589963,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.8463,
|
|
"mean_token_accuracy": 0.07354214936494827,
|
|
"num_tokens": 153121.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 8.101186609268188,
|
|
"epoch": 0.0800854244527496,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.7585,
|
|
"mean_token_accuracy": 0.07698194943368435,
|
|
"num_tokens": 164675.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 8.125961208343506,
|
|
"epoch": 0.08542445274959957,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.7999,
|
|
"mean_token_accuracy": 0.07594770528376102,
|
|
"num_tokens": 175146.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 8.141195964813232,
|
|
"epoch": 0.09076348104644955,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.8825,
|
|
"mean_token_accuracy": 0.06618092842400074,
|
|
"num_tokens": 186177.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 8.16831111907959,
|
|
"epoch": 0.09610250934329952,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.8295,
|
|
"mean_token_accuracy": 0.06966978162527085,
|
|
"num_tokens": 197143.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 8.187948417663574,
|
|
"epoch": 0.10144153764014949,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.8424,
|
|
"mean_token_accuracy": 0.06987790018320084,
|
|
"num_tokens": 207979.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 8.142891693115235,
|
|
"epoch": 0.10678056593699947,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.8131,
|
|
"mean_token_accuracy": 0.07337651327252388,
|
|
"num_tokens": 218733.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 8.133907985687255,
|
|
"epoch": 0.11211959423384944,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.8532,
|
|
"mean_token_accuracy": 0.06986705549061298,
|
|
"num_tokens": 228887.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 8.112241554260255,
|
|
"epoch": 0.11745862253069941,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.7902,
|
|
"mean_token_accuracy": 0.07219611443579196,
|
|
"num_tokens": 240289.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 8.137475204467773,
|
|
"epoch": 0.12279765082754938,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.8648,
|
|
"mean_token_accuracy": 0.07478626742959023,
|
|
"num_tokens": 250548.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 8.1454008102417,
|
|
"epoch": 0.12813667912439936,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.7908,
|
|
"mean_token_accuracy": 0.07535585910081863,
|
|
"num_tokens": 261669.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 8.10846242904663,
|
|
"epoch": 0.13347570742124934,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.8424,
|
|
"mean_token_accuracy": 0.07433099895715714,
|
|
"num_tokens": 272926.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 8.147141361236573,
|
|
"epoch": 0.1388147357180993,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.7477,
|
|
"mean_token_accuracy": 0.0750483900308609,
|
|
"num_tokens": 284173.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 8.172099399566651,
|
|
"epoch": 0.14415376401494928,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7941,
|
|
"mean_token_accuracy": 0.07234857454895974,
|
|
"num_tokens": 294850.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.106251430511474,
|
|
"epoch": 0.14949279231179927,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8398,
|
|
"mean_token_accuracy": 0.07133509777486324,
|
|
"num_tokens": 305727.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.116196441650391,
|
|
"epoch": 0.15483182060864922,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7629,
|
|
"mean_token_accuracy": 0.07706886641681195,
|
|
"num_tokens": 317243.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.17130641937256,
|
|
"epoch": 0.1601708489054992,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.8893,
|
|
"mean_token_accuracy": 0.07055514976382256,
|
|
"num_tokens": 328334.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.190022945404053,
|
|
"epoch": 0.16550987720234916,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8261,
|
|
"mean_token_accuracy": 0.06939826495945453,
|
|
"num_tokens": 339774.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.1759521484375,
|
|
"epoch": 0.17084890549919915,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8223,
|
|
"mean_token_accuracy": 0.07344263643026352,
|
|
"num_tokens": 350211.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.092726516723634,
|
|
"epoch": 0.1761879337960491,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7982,
|
|
"mean_token_accuracy": 0.07263314053416252,
|
|
"num_tokens": 360109.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.08326988220215,
|
|
"epoch": 0.1815269620928991,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.779,
|
|
"mean_token_accuracy": 0.07205527275800705,
|
|
"num_tokens": 370911.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.193898010253907,
|
|
"epoch": 0.18686599038974908,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8104,
|
|
"mean_token_accuracy": 0.0764952365309,
|
|
"num_tokens": 381808.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.056710386276245,
|
|
"epoch": 0.19220501868659903,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7494,
|
|
"mean_token_accuracy": 0.07683732137084007,
|
|
"num_tokens": 392175.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.197551059722901,
|
|
"epoch": 0.19754404698344902,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8564,
|
|
"mean_token_accuracy": 0.0671585526317358,
|
|
"num_tokens": 403261.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.099643325805664,
|
|
"epoch": 0.20288307528029897,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7407,
|
|
"mean_token_accuracy": 0.07576573640108109,
|
|
"num_tokens": 414992.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.113563442230225,
|
|
"epoch": 0.20822210357714896,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.754,
|
|
"mean_token_accuracy": 0.07667017802596092,
|
|
"num_tokens": 425316.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.137048149108887,
|
|
"epoch": 0.21356113187399894,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.8145,
|
|
"mean_token_accuracy": 0.07361620292067528,
|
|
"num_tokens": 437759.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.08598484992981,
|
|
"epoch": 0.2189001601708489,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7784,
|
|
"mean_token_accuracy": 0.07368727251887322,
|
|
"num_tokens": 448066.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.077975082397462,
|
|
"epoch": 0.22423918846769889,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7068,
|
|
"mean_token_accuracy": 0.07944998890161514,
|
|
"num_tokens": 458632.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.1232834815979,
|
|
"epoch": 0.22957821676454884,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7381,
|
|
"mean_token_accuracy": 0.07455957531929017,
|
|
"num_tokens": 469224.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.161263799667358,
|
|
"epoch": 0.23491724506139883,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.8058,
|
|
"mean_token_accuracy": 0.07331273853778839,
|
|
"num_tokens": 480492.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.1416805267334,
|
|
"epoch": 0.2402562733582488,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7529,
|
|
"mean_token_accuracy": 0.07843170017004013,
|
|
"num_tokens": 491325.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.065291547775269,
|
|
"epoch": 0.24559530165509877,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8025,
|
|
"mean_token_accuracy": 0.0766532089561224,
|
|
"num_tokens": 502240.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.098965644836426,
|
|
"epoch": 0.25093432995194875,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7488,
|
|
"mean_token_accuracy": 0.0799623742699623,
|
|
"num_tokens": 512710.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.122612571716308,
|
|
"epoch": 0.2562733582487987,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7845,
|
|
"mean_token_accuracy": 0.07602895051240921,
|
|
"num_tokens": 524023.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.073074436187744,
|
|
"epoch": 0.26161238654564867,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.7022,
|
|
"mean_token_accuracy": 0.07955551408231258,
|
|
"num_tokens": 534612.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.09334044456482,
|
|
"epoch": 0.2669514148424987,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7678,
|
|
"mean_token_accuracy": 0.07153690792620182,
|
|
"num_tokens": 544866.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.147175598144532,
|
|
"epoch": 0.27229044313934864,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.8087,
|
|
"mean_token_accuracy": 0.07087494693696499,
|
|
"num_tokens": 556091.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.130566596984863,
|
|
"epoch": 0.2776294714361986,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.805,
|
|
"mean_token_accuracy": 0.07419991269707679,
|
|
"num_tokens": 566946.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.10559663772583,
|
|
"epoch": 0.2829684997330486,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.6823,
|
|
"mean_token_accuracy": 0.07802616208791732,
|
|
"num_tokens": 577164.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.044942378997803,
|
|
"epoch": 0.28830752802989856,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.7173,
|
|
"mean_token_accuracy": 0.07408697754144669,
|
|
"num_tokens": 588744.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.085132789611816,
|
|
"epoch": 0.2936465563267485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.716,
|
|
"mean_token_accuracy": 0.07610422372817993,
|
|
"num_tokens": 600943.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.05778079032898,
|
|
"epoch": 0.29898558462359853,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.7647,
|
|
"mean_token_accuracy": 0.07883780151605606,
|
|
"num_tokens": 611864.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.074516010284423,
|
|
"epoch": 0.3043246129204485,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.7126,
|
|
"mean_token_accuracy": 0.07594608515501022,
|
|
"num_tokens": 623314.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.077810096740723,
|
|
"epoch": 0.30966364121729845,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7528,
|
|
"mean_token_accuracy": 0.07607153356075287,
|
|
"num_tokens": 633509.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.050742769241333,
|
|
"epoch": 0.3150026695141484,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6563,
|
|
"mean_token_accuracy": 0.08295750319957733,
|
|
"num_tokens": 644655.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.979042816162109,
|
|
"epoch": 0.3203416978109984,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.6899,
|
|
"mean_token_accuracy": 0.07570353075861931,
|
|
"num_tokens": 655940.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.179082679748536,
|
|
"epoch": 0.3256807261078484,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6518,
|
|
"mean_token_accuracy": 0.08190892487764359,
|
|
"num_tokens": 667074.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.127972841262817,
|
|
"epoch": 0.33101975440469833,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.7623,
|
|
"mean_token_accuracy": 0.07947569712996483,
|
|
"num_tokens": 677942.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.998423624038696,
|
|
"epoch": 0.33635878270154834,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6751,
|
|
"mean_token_accuracy": 0.08147123157978058,
|
|
"num_tokens": 689226.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.068508625030518,
|
|
"epoch": 0.3416978109983983,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.7416,
|
|
"mean_token_accuracy": 0.07528223097324371,
|
|
"num_tokens": 700830.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.10600996017456,
|
|
"epoch": 0.34703683929524826,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.7188,
|
|
"mean_token_accuracy": 0.07650677599012852,
|
|
"num_tokens": 712677.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.095576429367066,
|
|
"epoch": 0.3523758675920982,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.786,
|
|
"mean_token_accuracy": 0.07437327429652214,
|
|
"num_tokens": 723833.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.124511957168579,
|
|
"epoch": 0.3577148958889482,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.763,
|
|
"mean_token_accuracy": 0.07475610859692097,
|
|
"num_tokens": 735526.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.08799343109131,
|
|
"epoch": 0.3630539241857982,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.707,
|
|
"mean_token_accuracy": 0.0765246257185936,
|
|
"num_tokens": 745934.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.979455900192261,
|
|
"epoch": 0.36839295248264814,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.6603,
|
|
"mean_token_accuracy": 0.07649911046028138,
|
|
"num_tokens": 756693.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.125656652450562,
|
|
"epoch": 0.37373198077949815,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.7286,
|
|
"mean_token_accuracy": 0.07443214356899261,
|
|
"num_tokens": 767236.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.090236711502076,
|
|
"epoch": 0.3790710090763481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.6527,
|
|
"mean_token_accuracy": 0.0805337481200695,
|
|
"num_tokens": 778552.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.011722564697266,
|
|
"epoch": 0.38441003737319807,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.7639,
|
|
"mean_token_accuracy": 0.07575679123401642,
|
|
"num_tokens": 789615.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.122366762161255,
|
|
"epoch": 0.3897490656700481,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.7259,
|
|
"mean_token_accuracy": 0.07842210233211518,
|
|
"num_tokens": 799998.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.053548765182494,
|
|
"epoch": 0.39508809396689804,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6542,
|
|
"mean_token_accuracy": 0.0804023414850235,
|
|
"num_tokens": 810042.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.964968490600586,
|
|
"epoch": 0.400427122263748,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6593,
|
|
"mean_token_accuracy": 0.08147385269403458,
|
|
"num_tokens": 821692.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.106051301956176,
|
|
"epoch": 0.40576615056059795,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.6801,
|
|
"mean_token_accuracy": 0.0805871419608593,
|
|
"num_tokens": 832262.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.96937050819397,
|
|
"epoch": 0.41110517885744796,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.6426,
|
|
"mean_token_accuracy": 0.08224374130368232,
|
|
"num_tokens": 843006.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.042943239212036,
|
|
"epoch": 0.4164442071542979,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.6272,
|
|
"mean_token_accuracy": 0.07847281098365784,
|
|
"num_tokens": 853978.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.956211996078491,
|
|
"epoch": 0.4217832354511479,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.6663,
|
|
"mean_token_accuracy": 0.07781294696033,
|
|
"num_tokens": 864634.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.029766321182251,
|
|
"epoch": 0.4271222637479979,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.6238,
|
|
"mean_token_accuracy": 0.0834050565958023,
|
|
"num_tokens": 875379.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.130194091796875,
|
|
"epoch": 0.43246129204484784,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.671,
|
|
"mean_token_accuracy": 0.07772457972168922,
|
|
"num_tokens": 886428.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.950660228729248,
|
|
"epoch": 0.4378003203416978,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.6705,
|
|
"mean_token_accuracy": 0.08004693239927292,
|
|
"num_tokens": 896700.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.066575813293458,
|
|
"epoch": 0.44313934863854776,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.6757,
|
|
"mean_token_accuracy": 0.08255721405148506,
|
|
"num_tokens": 907478.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.987462520599365,
|
|
"epoch": 0.44847837693539777,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5464,
|
|
"mean_token_accuracy": 0.08664339035749435,
|
|
"num_tokens": 918170.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.865429925918579,
|
|
"epoch": 0.45381740523224773,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.6636,
|
|
"mean_token_accuracy": 0.08613021895289422,
|
|
"num_tokens": 928866.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.097207736968993,
|
|
"epoch": 0.4591564335290977,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.6311,
|
|
"mean_token_accuracy": 0.08184469342231751,
|
|
"num_tokens": 940395.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.960992765426636,
|
|
"epoch": 0.4644954618259477,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.6702,
|
|
"mean_token_accuracy": 0.07761614620685578,
|
|
"num_tokens": 950814.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.00200891494751,
|
|
"epoch": 0.46983449012279765,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.6605,
|
|
"mean_token_accuracy": 0.08137712515890598,
|
|
"num_tokens": 961658.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.125977849960327,
|
|
"epoch": 0.4751735184196476,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.6775,
|
|
"mean_token_accuracy": 0.0752656638622284,
|
|
"num_tokens": 974790.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.911575841903686,
|
|
"epoch": 0.4805125467164976,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.6076,
|
|
"mean_token_accuracy": 0.07721198052167892,
|
|
"num_tokens": 986599.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.994818162918091,
|
|
"epoch": 0.4858515750133476,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.6094,
|
|
"mean_token_accuracy": 0.07961493879556655,
|
|
"num_tokens": 997731.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.987026691436768,
|
|
"epoch": 0.49119060331019754,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.6809,
|
|
"mean_token_accuracy": 0.07975055277347565,
|
|
"num_tokens": 1009343.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.982935428619385,
|
|
"epoch": 0.4965296316070475,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.5893,
|
|
"mean_token_accuracy": 0.082920853048563,
|
|
"num_tokens": 1019878.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.984060430526734,
|
|
"epoch": 0.5018686599038975,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.5769,
|
|
"mean_token_accuracy": 0.08544681146740914,
|
|
"num_tokens": 1030551.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.922175931930542,
|
|
"epoch": 0.5072076882007475,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.6051,
|
|
"mean_token_accuracy": 0.08194510340690613,
|
|
"num_tokens": 1041432.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.9855893611907955,
|
|
"epoch": 0.5125467164975974,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.6037,
|
|
"mean_token_accuracy": 0.082750453799963,
|
|
"num_tokens": 1051769.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.929568576812744,
|
|
"epoch": 0.5178857447944474,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.5625,
|
|
"mean_token_accuracy": 0.08361941576004028,
|
|
"num_tokens": 1062334.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.949040317535401,
|
|
"epoch": 0.5232247730912973,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.5365,
|
|
"mean_token_accuracy": 0.08294312208890915,
|
|
"num_tokens": 1072792.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.850654411315918,
|
|
"epoch": 0.5285638013881474,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.4228,
|
|
"mean_token_accuracy": 0.08585703298449517,
|
|
"num_tokens": 1083642.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.871733045578003,
|
|
"epoch": 0.5339028296849974,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.5653,
|
|
"mean_token_accuracy": 0.08959369659423828,
|
|
"num_tokens": 1094454.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5339028296849974,
|
|
"eval_entropy": 7.677815706079657,
|
|
"eval_loss": 7.619736671447754,
|
|
"eval_mean_token_accuracy": 0.08080657599614813,
|
|
"eval_num_tokens": 1094454.0,
|
|
"eval_runtime": 2.3822,
|
|
"eval_samples_per_second": 1292.919,
|
|
"eval_steps_per_second": 161.615,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.980015087127685,
|
|
"epoch": 0.5392418579818473,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.6016,
|
|
"mean_token_accuracy": 0.07863722890615463,
|
|
"num_tokens": 1105068.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.854881715774536,
|
|
"epoch": 0.5445808862786973,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.5797,
|
|
"mean_token_accuracy": 0.08420702517032623,
|
|
"num_tokens": 1115668.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.96885986328125,
|
|
"epoch": 0.5499199145755472,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.5493,
|
|
"mean_token_accuracy": 0.08634284511208534,
|
|
"num_tokens": 1126742.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.003723287582398,
|
|
"epoch": 0.5552589428723972,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.6315,
|
|
"mean_token_accuracy": 0.07893695905804635,
|
|
"num_tokens": 1137351.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.812591028213501,
|
|
"epoch": 0.5605979711692471,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.5551,
|
|
"mean_token_accuracy": 0.08682225570082665,
|
|
"num_tokens": 1148389.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.921418476104736,
|
|
"epoch": 0.5659369994660972,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5444,
|
|
"mean_token_accuracy": 0.08775472566485405,
|
|
"num_tokens": 1158798.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.933099269866943,
|
|
"epoch": 0.5712760277629472,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.6203,
|
|
"mean_token_accuracy": 0.08136402703821659,
|
|
"num_tokens": 1169005.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.831244277954101,
|
|
"epoch": 0.5766150560597971,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.4835,
|
|
"mean_token_accuracy": 0.08241069503128529,
|
|
"num_tokens": 1180560.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.8498200416564945,
|
|
"epoch": 0.5819540843566471,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.4989,
|
|
"mean_token_accuracy": 0.08634271025657654,
|
|
"num_tokens": 1191162.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.867352056503296,
|
|
"epoch": 0.587293112653497,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.5098,
|
|
"mean_token_accuracy": 0.08468164131045341,
|
|
"num_tokens": 1202017.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.839261388778686,
|
|
"epoch": 0.592632140950347,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5011,
|
|
"mean_token_accuracy": 0.09654011353850364,
|
|
"num_tokens": 1213356.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.851275825500489,
|
|
"epoch": 0.5979711692471971,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.5056,
|
|
"mean_token_accuracy": 0.08764268457889557,
|
|
"num_tokens": 1224164.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.869636917114258,
|
|
"epoch": 0.603310197544047,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.5226,
|
|
"mean_token_accuracy": 0.08722739145159722,
|
|
"num_tokens": 1235483.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.813323211669922,
|
|
"epoch": 0.608649225840897,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.527,
|
|
"mean_token_accuracy": 0.08598990961909295,
|
|
"num_tokens": 1248013.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.9199892520904545,
|
|
"epoch": 0.6139882541377469,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.4867,
|
|
"mean_token_accuracy": 0.08738351389765739,
|
|
"num_tokens": 1258512.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.826328992843628,
|
|
"epoch": 0.6193272824345969,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.5097,
|
|
"mean_token_accuracy": 0.08399708047509194,
|
|
"num_tokens": 1269425.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.79866590499878,
|
|
"epoch": 0.6246663107314468,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.5051,
|
|
"mean_token_accuracy": 0.08688557520508766,
|
|
"num_tokens": 1280595.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.910846519470215,
|
|
"epoch": 0.6300053390282968,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.5351,
|
|
"mean_token_accuracy": 0.07902131229639053,
|
|
"num_tokens": 1290873.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.853626918792725,
|
|
"epoch": 0.6353443673251469,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.5742,
|
|
"mean_token_accuracy": 0.07897071465849877,
|
|
"num_tokens": 1303150.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.829471635818481,
|
|
"epoch": 0.6406833956219968,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.4643,
|
|
"mean_token_accuracy": 0.08364785388112068,
|
|
"num_tokens": 1314862.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.737740898132325,
|
|
"epoch": 0.6460224239188468,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.482,
|
|
"mean_token_accuracy": 0.08665243089199066,
|
|
"num_tokens": 1326052.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.872106599807739,
|
|
"epoch": 0.6513614522156967,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.4655,
|
|
"mean_token_accuracy": 0.08660055175423623,
|
|
"num_tokens": 1336847.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.802354240417481,
|
|
"epoch": 0.6567004805125467,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.4753,
|
|
"mean_token_accuracy": 0.09013021364808083,
|
|
"num_tokens": 1347611.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.81656436920166,
|
|
"epoch": 0.6620395088093967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.5065,
|
|
"mean_token_accuracy": 0.08921152874827384,
|
|
"num_tokens": 1358679.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.750188827514648,
|
|
"epoch": 0.6673785371062466,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.4782,
|
|
"mean_token_accuracy": 0.08896311670541764,
|
|
"num_tokens": 1369409.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.8263977527618405,
|
|
"epoch": 0.6727175654030967,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.4726,
|
|
"mean_token_accuracy": 0.09018750190734863,
|
|
"num_tokens": 1380380.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.7460150718688965,
|
|
"epoch": 0.6780565936999466,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.429,
|
|
"mean_token_accuracy": 0.08860143795609474,
|
|
"num_tokens": 1390885.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.830415964126587,
|
|
"epoch": 0.6833956219967966,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.09318191334605216,
|
|
"num_tokens": 1400960.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.7825994968414305,
|
|
"epoch": 0.6887346502936466,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.3978,
|
|
"mean_token_accuracy": 0.08887288495898246,
|
|
"num_tokens": 1411484.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.78870038986206,
|
|
"epoch": 0.6940736785904965,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4636,
|
|
"mean_token_accuracy": 0.08322765901684762,
|
|
"num_tokens": 1422931.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.799003076553345,
|
|
"epoch": 0.6994127068873465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.4119,
|
|
"mean_token_accuracy": 0.09033809676766395,
|
|
"num_tokens": 1434347.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.823387479782104,
|
|
"epoch": 0.7047517351841964,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.5046,
|
|
"mean_token_accuracy": 0.0883188046514988,
|
|
"num_tokens": 1445687.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.727633142471314,
|
|
"epoch": 0.7100907634810465,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.3436,
|
|
"mean_token_accuracy": 0.09381008222699165,
|
|
"num_tokens": 1456335.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.770741367340088,
|
|
"epoch": 0.7154297917778965,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.4241,
|
|
"mean_token_accuracy": 0.08320023119449615,
|
|
"num_tokens": 1467369.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.783315706253052,
|
|
"epoch": 0.7207688200747464,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.484,
|
|
"mean_token_accuracy": 0.08321434557437897,
|
|
"num_tokens": 1478734.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.785850381851196,
|
|
"epoch": 0.7261078483715964,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4656,
|
|
"mean_token_accuracy": 0.09263748377561569,
|
|
"num_tokens": 1489807.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.677286005020141,
|
|
"epoch": 0.7314468766684463,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.4125,
|
|
"mean_token_accuracy": 0.08424792289733887,
|
|
"num_tokens": 1501646.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.77900767326355,
|
|
"epoch": 0.7367859049652963,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.4372,
|
|
"mean_token_accuracy": 0.08848429471254349,
|
|
"num_tokens": 1511849.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.7723352909088135,
|
|
"epoch": 0.7421249332621462,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.4291,
|
|
"mean_token_accuracy": 0.08941564336419106,
|
|
"num_tokens": 1522183.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.7067298412323,
|
|
"epoch": 0.7474639615589963,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.4259,
|
|
"mean_token_accuracy": 0.08859878107905388,
|
|
"num_tokens": 1532819.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.78351788520813,
|
|
"epoch": 0.7528029898558463,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.3951,
|
|
"mean_token_accuracy": 0.09009868577122689,
|
|
"num_tokens": 1543909.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.756187677383423,
|
|
"epoch": 0.7581420181526962,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.4568,
|
|
"mean_token_accuracy": 0.08621637150645256,
|
|
"num_tokens": 1554147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.5803123950958256,
|
|
"epoch": 0.7634810464495462,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.345,
|
|
"mean_token_accuracy": 0.08573287092149258,
|
|
"num_tokens": 1565466.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.689743137359619,
|
|
"epoch": 0.7688200747463961,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.3258,
|
|
"mean_token_accuracy": 0.08579339385032654,
|
|
"num_tokens": 1577264.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.721196079254151,
|
|
"epoch": 0.7741591030432461,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.427,
|
|
"mean_token_accuracy": 0.08874332532286644,
|
|
"num_tokens": 1588222.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.620523738861084,
|
|
"epoch": 0.7794981313400962,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.3625,
|
|
"mean_token_accuracy": 0.08405865952372552,
|
|
"num_tokens": 1600609.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.796676826477051,
|
|
"epoch": 0.7848371596369461,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.4243,
|
|
"mean_token_accuracy": 0.08810800239443779,
|
|
"num_tokens": 1612032.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.686483478546142,
|
|
"epoch": 0.7901761879337961,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.4573,
|
|
"mean_token_accuracy": 0.0849332146346569,
|
|
"num_tokens": 1623228.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.690197467803955,
|
|
"epoch": 0.795515216230646,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.39,
|
|
"mean_token_accuracy": 0.09224065542221069,
|
|
"num_tokens": 1634293.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.710056304931641,
|
|
"epoch": 0.800854244527496,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.3702,
|
|
"mean_token_accuracy": 0.08933362364768982,
|
|
"num_tokens": 1645782.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.647915410995483,
|
|
"epoch": 0.8061932728243459,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3442,
|
|
"mean_token_accuracy": 0.09073336645960808,
|
|
"num_tokens": 1657060.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.582925653457641,
|
|
"epoch": 0.8115323011211959,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.3141,
|
|
"mean_token_accuracy": 0.09622450470924378,
|
|
"num_tokens": 1668017.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.726734733581543,
|
|
"epoch": 0.816871329418046,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.4015,
|
|
"mean_token_accuracy": 0.08468063101172447,
|
|
"num_tokens": 1678331.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.6575901985168455,
|
|
"epoch": 0.8222103577148959,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.3163,
|
|
"mean_token_accuracy": 0.09224411025643349,
|
|
"num_tokens": 1688719.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.626513385772705,
|
|
"epoch": 0.8275493860117459,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.3266,
|
|
"mean_token_accuracy": 0.08858170732855797,
|
|
"num_tokens": 1699028.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.703378677368164,
|
|
"epoch": 0.8328884143085958,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.3237,
|
|
"mean_token_accuracy": 0.08888528496026993,
|
|
"num_tokens": 1711122.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.628026008605957,
|
|
"epoch": 0.8382274426054458,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.3405,
|
|
"mean_token_accuracy": 0.09134237989783286,
|
|
"num_tokens": 1721694.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.588091897964477,
|
|
"epoch": 0.8435664709022958,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.363,
|
|
"mean_token_accuracy": 0.09324225187301635,
|
|
"num_tokens": 1732530.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.647649478912354,
|
|
"epoch": 0.8489054991991457,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.3364,
|
|
"mean_token_accuracy": 0.08565976619720458,
|
|
"num_tokens": 1742860.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.554535675048828,
|
|
"epoch": 0.8542445274959958,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.2888,
|
|
"mean_token_accuracy": 0.0925416223704815,
|
|
"num_tokens": 1752900.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.637857103347779,
|
|
"epoch": 0.8595835557928457,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.3142,
|
|
"mean_token_accuracy": 0.08696298599243164,
|
|
"num_tokens": 1763293.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.618868017196656,
|
|
"epoch": 0.8649225840896957,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.3186,
|
|
"mean_token_accuracy": 0.09385003075003624,
|
|
"num_tokens": 1773893.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.631073713302612,
|
|
"epoch": 0.8702616123865456,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.3823,
|
|
"mean_token_accuracy": 0.08830557242035866,
|
|
"num_tokens": 1785568.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.704895401000977,
|
|
"epoch": 0.8756006406833956,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.3165,
|
|
"mean_token_accuracy": 0.09338647127151489,
|
|
"num_tokens": 1797387.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.596003198623658,
|
|
"epoch": 0.8809396689802456,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.08846501111984253,
|
|
"num_tokens": 1808382.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.622437429428101,
|
|
"epoch": 0.8862786972770955,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.3526,
|
|
"mean_token_accuracy": 0.09341745972633361,
|
|
"num_tokens": 1818937.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.672019386291504,
|
|
"epoch": 0.8916177255739456,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.2736,
|
|
"mean_token_accuracy": 0.0895898312330246,
|
|
"num_tokens": 1830547.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.4498193740844725,
|
|
"epoch": 0.8969567538707955,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.2856,
|
|
"mean_token_accuracy": 0.09305597767233849,
|
|
"num_tokens": 1841569.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.639809083938599,
|
|
"epoch": 0.9022957821676455,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.287,
|
|
"mean_token_accuracy": 0.09668016657233239,
|
|
"num_tokens": 1852300.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.505797529220581,
|
|
"epoch": 0.9076348104644955,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.2914,
|
|
"mean_token_accuracy": 0.09635901600122451,
|
|
"num_tokens": 1863664.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.513959360122681,
|
|
"epoch": 0.9129738387613454,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.2111,
|
|
"mean_token_accuracy": 0.09574901312589645,
|
|
"num_tokens": 1873623.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.562733459472656,
|
|
"epoch": 0.9183128670581954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.3243,
|
|
"mean_token_accuracy": 0.09263002946972847,
|
|
"num_tokens": 1886092.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.581651830673218,
|
|
"epoch": 0.9236518953550453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.3213,
|
|
"mean_token_accuracy": 0.09076910391449929,
|
|
"num_tokens": 1897983.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.602070140838623,
|
|
"epoch": 0.9289909236518954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2949,
|
|
"mean_token_accuracy": 0.09758809804916382,
|
|
"num_tokens": 1908193.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.5159063816070555,
|
|
"epoch": 0.9343299519487454,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.2571,
|
|
"mean_token_accuracy": 0.09486114680767059,
|
|
"num_tokens": 1918628.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.578823614120483,
|
|
"epoch": 0.9396689802455953,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2993,
|
|
"mean_token_accuracy": 0.09509518221020699,
|
|
"num_tokens": 1929071.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.511225032806396,
|
|
"epoch": 0.9450080085424453,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.273,
|
|
"mean_token_accuracy": 0.09335564002394676,
|
|
"num_tokens": 1939567.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.581896018981934,
|
|
"epoch": 0.9503470368392952,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.3096,
|
|
"mean_token_accuracy": 0.09194429889321327,
|
|
"num_tokens": 1951165.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.629747295379639,
|
|
"epoch": 0.9556860651361452,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.2261,
|
|
"mean_token_accuracy": 0.09827369377017021,
|
|
"num_tokens": 1960855.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.5102427959442135,
|
|
"epoch": 0.9610250934329952,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.2183,
|
|
"mean_token_accuracy": 0.09632154554128647,
|
|
"num_tokens": 1971873.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.551190900802612,
|
|
"epoch": 0.9663641217298452,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.2724,
|
|
"mean_token_accuracy": 0.09511642530560493,
|
|
"num_tokens": 1983561.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.524335241317749,
|
|
"epoch": 0.9717031500266952,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.2505,
|
|
"mean_token_accuracy": 0.09372019693255425,
|
|
"num_tokens": 1994534.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.453141975402832,
|
|
"epoch": 0.9770421783235451,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.2176,
|
|
"mean_token_accuracy": 0.09602031484246254,
|
|
"num_tokens": 2005121.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.511305522918701,
|
|
"epoch": 0.9823812066203951,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.2205,
|
|
"mean_token_accuracy": 0.09739996567368507,
|
|
"num_tokens": 2016260.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.479723310470581,
|
|
"epoch": 0.987720234917245,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.1973,
|
|
"mean_token_accuracy": 0.09690724536776543,
|
|
"num_tokens": 2025705.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.499111175537109,
|
|
"epoch": 0.993059263214095,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.3585,
|
|
"mean_token_accuracy": 0.08967579677700996,
|
|
"num_tokens": 2036675.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.490612316131592,
|
|
"epoch": 0.9983982915109451,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.2061,
|
|
"mean_token_accuracy": 0.0968957595527172,
|
|
"num_tokens": 2047877.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.497582064734565,
|
|
"epoch": 1.0032034169781099,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.0981,
|
|
"mean_token_accuracy": 0.09796292831500371,
|
|
"num_tokens": 2057153.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.5237421035766605,
|
|
"epoch": 1.00854244527496,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.1893,
|
|
"mean_token_accuracy": 0.09472379684448243,
|
|
"num_tokens": 2068760.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.467091989517212,
|
|
"epoch": 1.01388147357181,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.0687,
|
|
"mean_token_accuracy": 0.10165433809161187,
|
|
"num_tokens": 2078671.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.499618721008301,
|
|
"epoch": 1.01922050186866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.0996,
|
|
"mean_token_accuracy": 0.10635028555989265,
|
|
"num_tokens": 2089581.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.431863689422608,
|
|
"epoch": 1.02455953016551,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.0515,
|
|
"mean_token_accuracy": 0.10709966346621513,
|
|
"num_tokens": 2100026.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.348581695556641,
|
|
"epoch": 1.02989855846236,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.0591,
|
|
"mean_token_accuracy": 0.10645127967000008,
|
|
"num_tokens": 2111838.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.479094982147217,
|
|
"epoch": 1.0352375867592098,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1476,
|
|
"mean_token_accuracy": 0.09466023072600364,
|
|
"num_tokens": 2123063.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.477510452270508,
|
|
"epoch": 1.0405766150560598,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.0467,
|
|
"mean_token_accuracy": 0.10368743538856506,
|
|
"num_tokens": 2134321.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.455883979797363,
|
|
"epoch": 1.0459156433529098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.0975,
|
|
"mean_token_accuracy": 0.10472378805279732,
|
|
"num_tokens": 2144216.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.359626293182373,
|
|
"epoch": 1.0512546716497597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.0779,
|
|
"mean_token_accuracy": 0.10107150450348854,
|
|
"num_tokens": 2155757.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.405184268951416,
|
|
"epoch": 1.0565936999466097,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.0661,
|
|
"mean_token_accuracy": 0.09650200307369232,
|
|
"num_tokens": 2166661.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.4126551151275635,
|
|
"epoch": 1.0619327282434596,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.0332,
|
|
"mean_token_accuracy": 0.10065687745809555,
|
|
"num_tokens": 2176796.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.3158660411834715,
|
|
"epoch": 1.0672717565403096,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.0972,
|
|
"mean_token_accuracy": 0.0997293807566166,
|
|
"num_tokens": 2186821.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0672717565403096,
|
|
"eval_entropy": 7.442174113880505,
|
|
"eval_loss": 7.287721157073975,
|
|
"eval_mean_token_accuracy": 0.09267236958560232,
|
|
"eval_num_tokens": 2186821.0,
|
|
"eval_runtime": 1.8223,
|
|
"eval_samples_per_second": 1690.14,
|
|
"eval_steps_per_second": 211.267,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.404880809783935,
|
|
"epoch": 1.0726107848371595,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999997458093253,
|
|
"loss": 7.0534,
|
|
"mean_token_accuracy": 0.10488023087382317,
|
|
"num_tokens": 2197953.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.367222642898559,
|
|
"epoch": 1.0779498131340097,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999987131606938,
|
|
"loss": 7.0678,
|
|
"mean_token_accuracy": 0.10656604394316674,
|
|
"num_tokens": 2208033.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.395871353149414,
|
|
"epoch": 1.0832888414308597,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999968861708311,
|
|
"loss": 7.0596,
|
|
"mean_token_accuracy": 0.0976042665541172,
|
|
"num_tokens": 2218010.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.474909734725952,
|
|
"epoch": 1.0886278697277096,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999942648461873,
|
|
"loss": 7.0793,
|
|
"mean_token_accuracy": 0.09963317140936852,
|
|
"num_tokens": 2229079.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.350576496124267,
|
|
"epoch": 1.0939668980245596,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999908491960169,
|
|
"loss": 7.0966,
|
|
"mean_token_accuracy": 0.09497827887535096,
|
|
"num_tokens": 2240333.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.298045682907104,
|
|
"epoch": 1.0993059263214096,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999866392323784,
|
|
"loss": 7.0422,
|
|
"mean_token_accuracy": 0.10446166545152664,
|
|
"num_tokens": 2251385.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.443157577514649,
|
|
"epoch": 1.1046449546182595,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999816349701349,
|
|
"loss": 7.0394,
|
|
"mean_token_accuracy": 0.10679379999637603,
|
|
"num_tokens": 2261890.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.239065551757813,
|
|
"epoch": 1.1099839829151095,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999758364269535,
|
|
"loss": 7.0638,
|
|
"mean_token_accuracy": 0.10125920474529267,
|
|
"num_tokens": 2272510.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.318096780776978,
|
|
"epoch": 1.1153230112119594,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999692436233055,
|
|
"loss": 7.0719,
|
|
"mean_token_accuracy": 0.09993241950869561,
|
|
"num_tokens": 2283392.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.3549572944641115,
|
|
"epoch": 1.1206620395088094,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999618565824665,
|
|
"loss": 6.9892,
|
|
"mean_token_accuracy": 0.10266446471214294,
|
|
"num_tokens": 2295155.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.384024906158447,
|
|
"epoch": 1.1260010678056593,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999536753305157,
|
|
"loss": 7.0816,
|
|
"mean_token_accuracy": 0.10478227883577347,
|
|
"num_tokens": 2307208.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.399112367630005,
|
|
"epoch": 1.1313400961025093,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999446998963364,
|
|
"loss": 7.0154,
|
|
"mean_token_accuracy": 0.09957597106695175,
|
|
"num_tokens": 2317480.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.360447549819947,
|
|
"epoch": 1.1366791243993593,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999349303116159,
|
|
"loss": 7.0993,
|
|
"mean_token_accuracy": 0.10178125277161598,
|
|
"num_tokens": 2328840.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.355537700653076,
|
|
"epoch": 1.1420181526962092,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999243666108448,
|
|
"loss": 7.0589,
|
|
"mean_token_accuracy": 0.09958812147378922,
|
|
"num_tokens": 2340450.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.372348737716675,
|
|
"epoch": 1.1473571809930592,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999130088313176,
|
|
"loss": 7.0946,
|
|
"mean_token_accuracy": 0.10345837995409965,
|
|
"num_tokens": 2351415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.324517059326172,
|
|
"epoch": 1.1526962092899091,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999008570131319,
|
|
"loss": 7.0889,
|
|
"mean_token_accuracy": 0.10542347058653831,
|
|
"num_tokens": 2362153.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.419645357131958,
|
|
"epoch": 1.1580352375867593,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998879111991889,
|
|
"loss": 7.0722,
|
|
"mean_token_accuracy": 0.10340167805552483,
|
|
"num_tokens": 2372352.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.280082893371582,
|
|
"epoch": 1.1633742658836093,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499874171435193,
|
|
"loss": 7.0409,
|
|
"mean_token_accuracy": 0.10494252145290375,
|
|
"num_tokens": 2382294.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.383985090255737,
|
|
"epoch": 1.1687132941804592,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998596377696511,
|
|
"loss": 7.0361,
|
|
"mean_token_accuracy": 0.10549432188272476,
|
|
"num_tokens": 2394054.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.230277729034424,
|
|
"epoch": 1.1740523224773092,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998443102538734,
|
|
"loss": 7.0257,
|
|
"mean_token_accuracy": 0.1024663895368576,
|
|
"num_tokens": 2405408.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.438461637496948,
|
|
"epoch": 1.1793913507741591,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998281889419725,
|
|
"loss": 7.08,
|
|
"mean_token_accuracy": 0.10087093561887742,
|
|
"num_tokens": 2416378.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.2362758159637455,
|
|
"epoch": 1.184730379071009,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998112738908636,
|
|
"loss": 6.9912,
|
|
"mean_token_accuracy": 0.1073272980749607,
|
|
"num_tokens": 2427016.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.2410563945770265,
|
|
"epoch": 1.190069407367859,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997935651602641,
|
|
"loss": 6.9181,
|
|
"mean_token_accuracy": 0.10832943990826607,
|
|
"num_tokens": 2437824.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.312191772460937,
|
|
"epoch": 1.195408435664709,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997750628126931,
|
|
"loss": 7.0515,
|
|
"mean_token_accuracy": 0.10108902752399444,
|
|
"num_tokens": 2448829.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.288640546798706,
|
|
"epoch": 1.200747463961559,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997557669134718,
|
|
"loss": 7.0153,
|
|
"mean_token_accuracy": 0.10415092334151269,
|
|
"num_tokens": 2459629.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.260274410247803,
|
|
"epoch": 1.206086492258409,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997356775307231,
|
|
"loss": 7.0824,
|
|
"mean_token_accuracy": 0.10463189780712127,
|
|
"num_tokens": 2471478.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.365083122253418,
|
|
"epoch": 1.2114255205552589,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499714794735371,
|
|
"loss": 6.9165,
|
|
"mean_token_accuracy": 0.11730484142899514,
|
|
"num_tokens": 2481573.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.196672677993774,
|
|
"epoch": 1.2167645488521088,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996931186011406,
|
|
"loss": 7.0444,
|
|
"mean_token_accuracy": 0.10180617272853851,
|
|
"num_tokens": 2493429.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.327607917785644,
|
|
"epoch": 1.222103577148959,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499670649204558,
|
|
"loss": 7.0127,
|
|
"mean_token_accuracy": 0.10604212060570717,
|
|
"num_tokens": 2503882.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.2563835144042965,
|
|
"epoch": 1.227442605445809,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996473866249496,
|
|
"loss": 6.9826,
|
|
"mean_token_accuracy": 0.10456614270806312,
|
|
"num_tokens": 2513978.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.318006181716919,
|
|
"epoch": 1.232781633742659,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996233309444424,
|
|
"loss": 7.0373,
|
|
"mean_token_accuracy": 0.10785176232457161,
|
|
"num_tokens": 2523806.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.281425189971924,
|
|
"epoch": 1.2381206620395089,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499598482247963,
|
|
"loss": 7.0411,
|
|
"mean_token_accuracy": 0.10471831485629082,
|
|
"num_tokens": 2533695.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.261021852493286,
|
|
"epoch": 1.2434596903363588,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995728406232378,
|
|
"loss": 6.9673,
|
|
"mean_token_accuracy": 0.10838826149702072,
|
|
"num_tokens": 2544315.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.233233118057251,
|
|
"epoch": 1.2487987186332088,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499546406160793,
|
|
"loss": 6.9647,
|
|
"mean_token_accuracy": 0.10495424494147301,
|
|
"num_tokens": 2555985.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.298960876464844,
|
|
"epoch": 1.2541377469300587,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995191789539531,
|
|
"loss": 7.043,
|
|
"mean_token_accuracy": 0.09680852070450782,
|
|
"num_tokens": 2567472.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.300242137908936,
|
|
"epoch": 1.2594767752269087,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994911590988421,
|
|
"loss": 7.0309,
|
|
"mean_token_accuracy": 0.10832907631993294,
|
|
"num_tokens": 2578030.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.305524635314941,
|
|
"epoch": 1.2648158035237587,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994623466943816,
|
|
"loss": 7.0479,
|
|
"mean_token_accuracy": 0.10340372696518899,
|
|
"num_tokens": 2589925.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.2693243503570555,
|
|
"epoch": 1.2701548318206086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004994327418422922,
|
|
"loss": 6.9683,
|
|
"mean_token_accuracy": 0.10609261021018028,
|
|
"num_tokens": 2601825.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.242023992538452,
|
|
"epoch": 1.2754938601174586,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994023446470912,
|
|
"loss": 7.0,
|
|
"mean_token_accuracy": 0.1082241028547287,
|
|
"num_tokens": 2612726.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.156720399856567,
|
|
"epoch": 1.2808328884143085,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004993711552160936,
|
|
"loss": 6.9354,
|
|
"mean_token_accuracy": 0.10554772540926934,
|
|
"num_tokens": 2623753.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.261723423004151,
|
|
"epoch": 1.2861719167111585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993391736594117,
|
|
"loss": 6.9408,
|
|
"mean_token_accuracy": 0.1137481302022934,
|
|
"num_tokens": 2635275.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.308083820343017,
|
|
"epoch": 1.2915109450080084,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993064000899538,
|
|
"loss": 6.9987,
|
|
"mean_token_accuracy": 0.11398440822958947,
|
|
"num_tokens": 2646329.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.151524448394776,
|
|
"epoch": 1.2968499733048584,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004992728346234246,
|
|
"loss": 7.0185,
|
|
"mean_token_accuracy": 0.10763908922672272,
|
|
"num_tokens": 2657629.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.314895677566528,
|
|
"epoch": 1.3021890016017084,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992384773783244,
|
|
"loss": 7.008,
|
|
"mean_token_accuracy": 0.10860289186239243,
|
|
"num_tokens": 2667746.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.13032865524292,
|
|
"epoch": 1.3075280298985585,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004992033284759491,
|
|
"loss": 6.9652,
|
|
"mean_token_accuracy": 0.11226292178034783,
|
|
"num_tokens": 2678157.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.2998748302459715,
|
|
"epoch": 1.3128670581954085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991673880403891,
|
|
"loss": 7.0113,
|
|
"mean_token_accuracy": 0.09815131574869156,
|
|
"num_tokens": 2689319.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.274382066726685,
|
|
"epoch": 1.3182060864922585,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991306561985294,
|
|
"loss": 6.9558,
|
|
"mean_token_accuracy": 0.10922536626458168,
|
|
"num_tokens": 2699718.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.224872779846192,
|
|
"epoch": 1.3235451147891084,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990931330800492,
|
|
"loss": 6.9992,
|
|
"mean_token_accuracy": 0.107916821539402,
|
|
"num_tokens": 2710888.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.232290554046631,
|
|
"epoch": 1.3288841430859584,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990548188174211,
|
|
"loss": 6.9229,
|
|
"mean_token_accuracy": 0.10669831410050393,
|
|
"num_tokens": 2722024.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.180148696899414,
|
|
"epoch": 1.3342231713828083,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990157135459105,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.10785769298672676,
|
|
"num_tokens": 2731780.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.226172161102295,
|
|
"epoch": 1.3395621996796583,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989758174035758,
|
|
"loss": 7.0046,
|
|
"mean_token_accuracy": 0.10561225786805153,
|
|
"num_tokens": 2742524.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.261577129364014,
|
|
"epoch": 1.3449012279765082,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004989351305312676,
|
|
"loss": 6.9437,
|
|
"mean_token_accuracy": 0.11086614355444908,
|
|
"num_tokens": 2752077.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.192623233795166,
|
|
"epoch": 1.3502402562733582,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004988936530726276,
|
|
"loss": 6.9629,
|
|
"mean_token_accuracy": 0.10704695656895638,
|
|
"num_tokens": 2763625.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.103010416030884,
|
|
"epoch": 1.3555792845702082,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988513851740889,
|
|
"loss": 6.9389,
|
|
"mean_token_accuracy": 0.11248037964105606,
|
|
"num_tokens": 2775262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.300057458877563,
|
|
"epoch": 1.3609183128670583,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004988083269848753,
|
|
"loss": 6.9964,
|
|
"mean_token_accuracy": 0.10848051384091377,
|
|
"num_tokens": 2786449.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.15974440574646,
|
|
"epoch": 1.3662573411639083,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004987644786570003,
|
|
"loss": 6.9879,
|
|
"mean_token_accuracy": 0.1025148332118988,
|
|
"num_tokens": 2796891.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.209011840820312,
|
|
"epoch": 1.3715963694607582,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004987198403452674,
|
|
"loss": 6.9917,
|
|
"mean_token_accuracy": 0.10826320275664329,
|
|
"num_tokens": 2806926.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.288408470153809,
|
|
"epoch": 1.3769353977576082,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986744122072688,
|
|
"loss": 6.8986,
|
|
"mean_token_accuracy": 0.10854956284165382,
|
|
"num_tokens": 2817750.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.049510908126831,
|
|
"epoch": 1.3822744260544582,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004986281944033851,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.11455944776535035,
|
|
"num_tokens": 2827770.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.274826765060425,
|
|
"epoch": 1.3876134543513081,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004985811870967848,
|
|
"loss": 6.9575,
|
|
"mean_token_accuracy": 0.1100310817360878,
|
|
"num_tokens": 2839583.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.212510490417481,
|
|
"epoch": 1.392952482648158,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.9606,
|
|
"mean_token_accuracy": 0.1118274174630642,
|
|
"num_tokens": 2850621.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.1082500457763675,
|
|
"epoch": 1.398291510945008,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984848046420445,
|
|
"loss": 6.959,
|
|
"mean_token_accuracy": 0.10316182821989059,
|
|
"num_tokens": 2862287.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.263628625869751,
|
|
"epoch": 1.403630539241858,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004984354298341756,
|
|
"loss": 6.9509,
|
|
"mean_token_accuracy": 0.11599895805120468,
|
|
"num_tokens": 2871769.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.160809659957886,
|
|
"epoch": 1.408969567538708,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004983852662041312,
|
|
"loss": 6.9293,
|
|
"mean_token_accuracy": 0.10901639312505722,
|
|
"num_tokens": 2882590.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.240030336380005,
|
|
"epoch": 1.414308595835558,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00049833431392901,
|
|
"loss": 6.9903,
|
|
"mean_token_accuracy": 0.099904265999794,
|
|
"num_tokens": 2893903.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.11286301612854,
|
|
"epoch": 1.4196476241324079,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004982825731886956,
|
|
"loss": 6.898,
|
|
"mean_token_accuracy": 0.10848561599850655,
|
|
"num_tokens": 2904532.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.1621256351470945,
|
|
"epoch": 1.4249866524292578,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004982300441658546,
|
|
"loss": 6.891,
|
|
"mean_token_accuracy": 0.10317038744688034,
|
|
"num_tokens": 2916340.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.138517284393311,
|
|
"epoch": 1.4303256807261078,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004981767270459367,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.10857687294483184,
|
|
"num_tokens": 2926424.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.2230969905853275,
|
|
"epoch": 1.4356647090229577,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981226220171744,
|
|
"loss": 6.962,
|
|
"mean_token_accuracy": 0.1114728294312954,
|
|
"num_tokens": 2937620.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.150681591033935,
|
|
"epoch": 1.4410037373198077,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004980677292705811,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.11602405309677125,
|
|
"num_tokens": 2948274.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.103616523742676,
|
|
"epoch": 1.4463427656166576,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004980120489999519,
|
|
"loss": 6.9584,
|
|
"mean_token_accuracy": 0.10770896971225738,
|
|
"num_tokens": 2958986.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.292509651184082,
|
|
"epoch": 1.4516817939135078,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979555814018617,
|
|
"loss": 6.9853,
|
|
"mean_token_accuracy": 0.1034561313688755,
|
|
"num_tokens": 2970933.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.148707580566406,
|
|
"epoch": 1.4570208222103578,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497898326675665,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.10509125217795372,
|
|
"num_tokens": 2981197.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.232423114776611,
|
|
"epoch": 1.4623598505072077,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004978402850234956,
|
|
"loss": 6.9586,
|
|
"mean_token_accuracy": 0.11153719946742058,
|
|
"num_tokens": 2993732.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.131136846542359,
|
|
"epoch": 1.4676988788040577,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004977814566502651,
|
|
"loss": 6.8896,
|
|
"mean_token_accuracy": 0.11299286112189293,
|
|
"num_tokens": 3004617.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.089672565460205,
|
|
"epoch": 1.4730379071009077,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004977218417636629,
|
|
"loss": 6.8467,
|
|
"mean_token_accuracy": 0.11408524513244629,
|
|
"num_tokens": 3015058.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.131562614440918,
|
|
"epoch": 1.4783769353977576,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004976614405741546,
|
|
"loss": 6.838,
|
|
"mean_token_accuracy": 0.11239745616912841,
|
|
"num_tokens": 3025995.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.124327945709228,
|
|
"epoch": 1.4837159636946076,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004976002532949823,
|
|
"loss": 6.8728,
|
|
"mean_token_accuracy": 0.11653813868761062,
|
|
"num_tokens": 3036060.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.0725609302520756,
|
|
"epoch": 1.4890549919914575,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004975382801421631,
|
|
"loss": 6.8935,
|
|
"mean_token_accuracy": 0.11659580916166305,
|
|
"num_tokens": 3047012.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.082496929168701,
|
|
"epoch": 1.4943940202883075,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974755213344886,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.11371453702449799,
|
|
"num_tokens": 3058186.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.047554445266724,
|
|
"epoch": 1.4997330485851574,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004974119770935242,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.11552045941352844,
|
|
"num_tokens": 3068566.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.140801477432251,
|
|
"epoch": 1.5050720768820076,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004973476476436081,
|
|
"loss": 6.8548,
|
|
"mean_token_accuracy": 0.1138184629380703,
|
|
"num_tokens": 3079588.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.0329976081848145,
|
|
"epoch": 1.5104111051788576,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972825332118505,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.11353544220328331,
|
|
"num_tokens": 3092376.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.064114332199097,
|
|
"epoch": 1.5157501334757075,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972166340281331,
|
|
"loss": 6.8672,
|
|
"mean_token_accuracy": 0.11258345693349839,
|
|
"num_tokens": 3103439.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.15282621383667,
|
|
"epoch": 1.5210891617725575,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004971499503251083,
|
|
"loss": 6.928,
|
|
"mean_token_accuracy": 0.10791781321167945,
|
|
"num_tokens": 3115007.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.142194128036499,
|
|
"epoch": 1.5264281900694074,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004970824823381976,
|
|
"loss": 6.8976,
|
|
"mean_token_accuracy": 0.10874230861663818,
|
|
"num_tokens": 3125632.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.102570247650147,
|
|
"epoch": 1.5317672183662574,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497014230305592,
|
|
"loss": 6.8638,
|
|
"mean_token_accuracy": 0.11153942048549652,
|
|
"num_tokens": 3135825.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.117893218994141,
|
|
"epoch": 1.5371062466631074,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004969451944682501,
|
|
"loss": 6.8776,
|
|
"mean_token_accuracy": 0.10968513563275337,
|
|
"num_tokens": 3147689.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.995888614654541,
|
|
"epoch": 1.5424452749599573,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004968753750698978,
|
|
"loss": 6.825,
|
|
"mean_token_accuracy": 0.1107549749314785,
|
|
"num_tokens": 3158754.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.188800191879272,
|
|
"epoch": 1.5477843032568073,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004968047723570273,
|
|
"loss": 6.8774,
|
|
"mean_token_accuracy": 0.10438808798789978,
|
|
"num_tokens": 3169332.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.127320146560669,
|
|
"epoch": 1.5531233315536572,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004967333865788961,
|
|
"loss": 6.9489,
|
|
"mean_token_accuracy": 0.11461934968829154,
|
|
"num_tokens": 3180761.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.155689430236817,
|
|
"epoch": 1.5584623598505072,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004966612179875267,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.10985235571861267,
|
|
"num_tokens": 3191635.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.092573738098144,
|
|
"epoch": 1.5638013881473571,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004965882668377048,
|
|
"loss": 6.875,
|
|
"mean_token_accuracy": 0.10863233283162117,
|
|
"num_tokens": 3202452.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.097639560699463,
|
|
"epoch": 1.569140416444207,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004965145333869789,
|
|
"loss": 6.8113,
|
|
"mean_token_accuracy": 0.11319924965500831,
|
|
"num_tokens": 3213298.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.036502075195313,
|
|
"epoch": 1.574479444741057,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004964400178956595,
|
|
"loss": 6.839,
|
|
"mean_token_accuracy": 0.11598326861858368,
|
|
"num_tokens": 3224314.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.094682455062866,
|
|
"epoch": 1.579818473037907,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004963647206268181,
|
|
"loss": 6.8703,
|
|
"mean_token_accuracy": 0.11335297301411629,
|
|
"num_tokens": 3235660.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.1418054580688475,
|
|
"epoch": 1.585157501334757,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000496288641846286,
|
|
"loss": 6.9191,
|
|
"mean_token_accuracy": 0.10401919782161713,
|
|
"num_tokens": 3246856.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.106872415542602,
|
|
"epoch": 1.590496529631607,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004962117818226538,
|
|
"loss": 6.9676,
|
|
"mean_token_accuracy": 0.10796281695365906,
|
|
"num_tokens": 3258692.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.085078763961792,
|
|
"epoch": 1.5958355579284569,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00049613414082727,
|
|
"loss": 6.8555,
|
|
"mean_token_accuracy": 0.10882437452673913,
|
|
"num_tokens": 3270577.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.117661190032959,
|
|
"epoch": 1.6011745862253068,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004960557191342405,
|
|
"loss": 6.7828,
|
|
"mean_token_accuracy": 0.11544768214225769,
|
|
"num_tokens": 3281262.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6011745862253068,
|
|
"eval_entropy": 6.949690158645828,
|
|
"eval_loss": 7.0078125,
|
|
"eval_mean_token_accuracy": 0.10770162927446428,
|
|
"eval_num_tokens": 3281262.0,
|
|
"eval_runtime": 1.8094,
|
|
"eval_samples_per_second": 1702.193,
|
|
"eval_steps_per_second": 212.774,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.052781963348389,
|
|
"epoch": 1.606513614522157,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000495976517020427,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.11890260726213456,
|
|
"num_tokens": 3291552.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.097465991973877,
|
|
"epoch": 1.611852642819007,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004958965347654469,
|
|
"loss": 6.8993,
|
|
"mean_token_accuracy": 0.11008531674742698,
|
|
"num_tokens": 3301977.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.056075668334961,
|
|
"epoch": 1.617191671115857,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004958157726516715,
|
|
"loss": 6.8217,
|
|
"mean_token_accuracy": 0.11723290756344795,
|
|
"num_tokens": 3312783.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.090235471725464,
|
|
"epoch": 1.622530699412707,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004957342309642256,
|
|
"loss": 6.8343,
|
|
"mean_token_accuracy": 0.11337208822369575,
|
|
"num_tokens": 3323895.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.960222244262695,
|
|
"epoch": 1.6278697277095568,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004956519099909857,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.11581635996699333,
|
|
"num_tokens": 3334103.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.116361856460571,
|
|
"epoch": 1.6332087560064068,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004955688100225803,
|
|
"loss": 6.8234,
|
|
"mean_token_accuracy": 0.11667298898100853,
|
|
"num_tokens": 3346279.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.90750412940979,
|
|
"epoch": 1.6385477843032568,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004954849313523875,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.11140712723135948,
|
|
"num_tokens": 3357582.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.171404266357422,
|
|
"epoch": 1.643886812600107,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004954002742765347,
|
|
"loss": 6.8133,
|
|
"mean_token_accuracy": 0.11421637535095215,
|
|
"num_tokens": 3369297.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.962291955947876,
|
|
"epoch": 1.649225840896957,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004953148390938973,
|
|
"loss": 6.9113,
|
|
"mean_token_accuracy": 0.11368932947516441,
|
|
"num_tokens": 3382430.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.1052000522613525,
|
|
"epoch": 1.6545648691938069,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000495228626106098,
|
|
"loss": 6.886,
|
|
"mean_token_accuracy": 0.12085917294025421,
|
|
"num_tokens": 3393809.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.020896577835083,
|
|
"epoch": 1.6599038974906568,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004951416356175053,
|
|
"loss": 6.8681,
|
|
"mean_token_accuracy": 0.1101914569735527,
|
|
"num_tokens": 3405164.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.101365995407105,
|
|
"epoch": 1.6652429257875068,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004950538679352327,
|
|
"loss": 6.8741,
|
|
"mean_token_accuracy": 0.1086366482079029,
|
|
"num_tokens": 3416091.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.135610580444336,
|
|
"epoch": 1.6705819540843567,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004949653233691373,
|
|
"loss": 6.9041,
|
|
"mean_token_accuracy": 0.11493342369794846,
|
|
"num_tokens": 3427675.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.049169683456421,
|
|
"epoch": 1.6759209823812067,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004948760022318194,
|
|
"loss": 6.7869,
|
|
"mean_token_accuracy": 0.11913650557398796,
|
|
"num_tokens": 3438119.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.976612186431884,
|
|
"epoch": 1.6812600106780566,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004947859048386205,
|
|
"loss": 6.8392,
|
|
"mean_token_accuracy": 0.10995432883501052,
|
|
"num_tokens": 3449151.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.030420732498169,
|
|
"epoch": 1.6865990389749066,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004946950315076225,
|
|
"loss": 6.7241,
|
|
"mean_token_accuracy": 0.1148534007370472,
|
|
"num_tokens": 3459697.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.9406853199005125,
|
|
"epoch": 1.6919380672717566,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004946033825596474,
|
|
"loss": 6.8019,
|
|
"mean_token_accuracy": 0.1164799764752388,
|
|
"num_tokens": 3470409.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.096703147888183,
|
|
"epoch": 1.6972770955686065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945109583182548,
|
|
"loss": 6.818,
|
|
"mean_token_accuracy": 0.11084006652235985,
|
|
"num_tokens": 3481515.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.999007701873779,
|
|
"epoch": 1.7026161238654565,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004944177591097415,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.11739434972405434,
|
|
"num_tokens": 3492683.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.013306427001953,
|
|
"epoch": 1.7079551521623064,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943237852631404,
|
|
"loss": 6.793,
|
|
"mean_token_accuracy": 0.11535104885697364,
|
|
"num_tokens": 3503067.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.023929929733276,
|
|
"epoch": 1.7132941804591564,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004942290371102193,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.118143829703331,
|
|
"num_tokens": 3513611.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.093850612640381,
|
|
"epoch": 1.7186332087560063,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004941335149854794,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.11167100667953492,
|
|
"num_tokens": 3523802.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.002861070632934,
|
|
"epoch": 1.7239722370528563,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004940372192261546,
|
|
"loss": 6.7916,
|
|
"mean_token_accuracy": 0.11050481945276261,
|
|
"num_tokens": 3534764.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.046298027038574,
|
|
"epoch": 1.7293112653497063,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004939401501722098,
|
|
"loss": 6.8578,
|
|
"mean_token_accuracy": 0.10906066969037057,
|
|
"num_tokens": 3544852.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.963241100311279,
|
|
"epoch": 1.7346502936465562,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004938423081663401,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.11659257933497429,
|
|
"num_tokens": 3555470.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.011042356491089,
|
|
"epoch": 1.7399893219434062,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004937436935539695,
|
|
"loss": 6.8583,
|
|
"mean_token_accuracy": 0.10600545033812522,
|
|
"num_tokens": 3567497.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.062386178970337,
|
|
"epoch": 1.7453283502402561,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004936443066832494,
|
|
"loss": 6.825,
|
|
"mean_token_accuracy": 0.11599826514720916,
|
|
"num_tokens": 3578354.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.932935047149658,
|
|
"epoch": 1.7506673785371063,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004935441479050579,
|
|
"loss": 6.7342,
|
|
"mean_token_accuracy": 0.1210530936717987,
|
|
"num_tokens": 3589369.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.086917066574097,
|
|
"epoch": 1.7560064068339563,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004934432175729981,
|
|
"loss": 6.7596,
|
|
"mean_token_accuracy": 0.11764036342501641,
|
|
"num_tokens": 3600396.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.000305032730102,
|
|
"epoch": 1.7613454351308062,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000493341516043397,
|
|
"loss": 6.8199,
|
|
"mean_token_accuracy": 0.11525925770401954,
|
|
"num_tokens": 3610670.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.981289863586426,
|
|
"epoch": 1.7666844634276562,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004932390436753043,
|
|
"loss": 6.7769,
|
|
"mean_token_accuracy": 0.11980077475309373,
|
|
"num_tokens": 3620821.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.0027588367462155,
|
|
"epoch": 1.7720234917245061,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004931358008304911,
|
|
"loss": 6.7683,
|
|
"mean_token_accuracy": 0.11936180293560028,
|
|
"num_tokens": 3631469.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.988579654693604,
|
|
"epoch": 1.777362520021356,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004930317878734486,
|
|
"loss": 6.7774,
|
|
"mean_token_accuracy": 0.11964918524026871,
|
|
"num_tokens": 3641605.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.027560663223267,
|
|
"epoch": 1.782701548318206,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004929270051713866,
|
|
"loss": 6.8064,
|
|
"mean_token_accuracy": 0.11727471053600311,
|
|
"num_tokens": 3652107.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.005648851394653,
|
|
"epoch": 1.7880405766150562,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004928214530942331,
|
|
"loss": 6.8327,
|
|
"mean_token_accuracy": 0.11391589567065238,
|
|
"num_tokens": 3663399.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.932778549194336,
|
|
"epoch": 1.7933796049119062,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004927151320146313,
|
|
"loss": 6.8056,
|
|
"mean_token_accuracy": 0.11577988713979721,
|
|
"num_tokens": 3673638.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.997147560119629,
|
|
"epoch": 1.7987186332087561,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 6.7687,
|
|
"mean_token_accuracy": 0.1166080392897129,
|
|
"num_tokens": 3684459.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.988068914413452,
|
|
"epoch": 1.804057661505606,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004925001843522323,
|
|
"loss": 6.8131,
|
|
"mean_token_accuracy": 0.11480777710676193,
|
|
"num_tokens": 3696113.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.034640789031982,
|
|
"epoch": 1.809396689802456,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004923915585282916,
|
|
"loss": 6.7798,
|
|
"mean_token_accuracy": 0.11368410959839821,
|
|
"num_tokens": 3707345.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.962531518936157,
|
|
"epoch": 1.814735718099306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004922821652196136,
|
|
"loss": 6.6051,
|
|
"mean_token_accuracy": 0.12565283328294755,
|
|
"num_tokens": 3717573.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.9115925312042235,
|
|
"epoch": 1.820074746396156,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004921720048124034,
|
|
"loss": 6.7772,
|
|
"mean_token_accuracy": 0.11985760256648063,
|
|
"num_tokens": 3726658.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.99907660484314,
|
|
"epoch": 1.825413774693006,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004920610776955742,
|
|
"loss": 6.8131,
|
|
"mean_token_accuracy": 0.11796022430062295,
|
|
"num_tokens": 3737769.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.914171552658081,
|
|
"epoch": 1.8307528029898559,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004919493842607456,
|
|
"loss": 6.7338,
|
|
"mean_token_accuracy": 0.11815480217337608,
|
|
"num_tokens": 3748608.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.033916187286377,
|
|
"epoch": 1.8360918312867058,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004918369249022433,
|
|
"loss": 6.7885,
|
|
"mean_token_accuracy": 0.11544626578688622,
|
|
"num_tokens": 3759063.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.005947017669678,
|
|
"epoch": 1.8414308595835558,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004917237000170966,
|
|
"loss": 6.7356,
|
|
"mean_token_accuracy": 0.11510898619890213,
|
|
"num_tokens": 3769622.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.9657557010650635,
|
|
"epoch": 1.8467698878804057,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004916097100050375,
|
|
"loss": 6.7251,
|
|
"mean_token_accuracy": 0.12068109884858132,
|
|
"num_tokens": 3780587.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.938025617599488,
|
|
"epoch": 1.8521089161772557,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004914949552684992,
|
|
"loss": 6.8035,
|
|
"mean_token_accuracy": 0.11396916881203652,
|
|
"num_tokens": 3792386.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.017823553085327,
|
|
"epoch": 1.8574479444741057,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000491379436212615,
|
|
"loss": 6.8248,
|
|
"mean_token_accuracy": 0.11695968508720397,
|
|
"num_tokens": 3803820.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.960682201385498,
|
|
"epoch": 1.8627869727709556,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000491263153245216,
|
|
"loss": 6.7841,
|
|
"mean_token_accuracy": 0.11130457371473312,
|
|
"num_tokens": 3815187.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.0006937980651855,
|
|
"epoch": 1.8681260010678056,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004911461067768309,
|
|
"loss": 6.7962,
|
|
"mean_token_accuracy": 0.11658177971839905,
|
|
"num_tokens": 3826933.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.948520612716675,
|
|
"epoch": 1.8734650293646555,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004910282972206834,
|
|
"loss": 6.6597,
|
|
"mean_token_accuracy": 0.1199097216129303,
|
|
"num_tokens": 3836934.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.905463218688965,
|
|
"epoch": 1.8788040576615055,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004909097249926913,
|
|
"loss": 6.7612,
|
|
"mean_token_accuracy": 0.11695051416754723,
|
|
"num_tokens": 3848662.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.947757577896118,
|
|
"epoch": 1.8841430859583554,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004907903905114652,
|
|
"loss": 6.629,
|
|
"mean_token_accuracy": 0.12088115662336349,
|
|
"num_tokens": 3860052.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.922029829025268,
|
|
"epoch": 1.8894821142552054,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004906702941983066,
|
|
"loss": 6.7547,
|
|
"mean_token_accuracy": 0.11512938812375069,
|
|
"num_tokens": 3871867.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.998465299606323,
|
|
"epoch": 1.8948211425520554,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004905494364772065,
|
|
"loss": 6.8035,
|
|
"mean_token_accuracy": 0.11333657875657081,
|
|
"num_tokens": 3883612.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.89701657295227,
|
|
"epoch": 1.9001601708489055,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004904278177748441,
|
|
"loss": 6.7159,
|
|
"mean_token_accuracy": 0.12031620517373084,
|
|
"num_tokens": 3895520.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.97585654258728,
|
|
"epoch": 1.9054991991457555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004903054385205854,
|
|
"loss": 6.7116,
|
|
"mean_token_accuracy": 0.11240454241633416,
|
|
"num_tokens": 3905898.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.992402124404907,
|
|
"epoch": 1.9108382274426055,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004901822991464811,
|
|
"loss": 6.7691,
|
|
"mean_token_accuracy": 0.11295585930347443,
|
|
"num_tokens": 3916339.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.9712841510772705,
|
|
"epoch": 1.9161772557394554,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004900584000872655,
|
|
"loss": 6.7394,
|
|
"mean_token_accuracy": 0.11573940142989159,
|
|
"num_tokens": 3927131.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.857951068878174,
|
|
"epoch": 1.9215162840363054,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004899337417803551,
|
|
"loss": 6.7306,
|
|
"mean_token_accuracy": 0.12320442199707031,
|
|
"num_tokens": 3938415.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.0113670349121096,
|
|
"epoch": 1.9268553123331553,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000489808324665847,
|
|
"loss": 6.7351,
|
|
"mean_token_accuracy": 0.12133737727999687,
|
|
"num_tokens": 3949700.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.952113294601441,
|
|
"epoch": 1.9321943406300055,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004896821491865167,
|
|
"loss": 6.7885,
|
|
"mean_token_accuracy": 0.117950689047575,
|
|
"num_tokens": 3961879.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.012113094329834,
|
|
"epoch": 1.9375333689268555,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004895552157878174,
|
|
"loss": 6.765,
|
|
"mean_token_accuracy": 0.12073935866355896,
|
|
"num_tokens": 3972903.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.899973821640015,
|
|
"epoch": 1.9428723972237054,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004894275249178782,
|
|
"loss": 6.7555,
|
|
"mean_token_accuracy": 0.12310554385185242,
|
|
"num_tokens": 3984432.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.9715163230896,
|
|
"epoch": 1.9482114255205554,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004892990770275021,
|
|
"loss": 6.806,
|
|
"mean_token_accuracy": 0.10806804299354553,
|
|
"num_tokens": 3996998.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.985030603408814,
|
|
"epoch": 1.9535504538174053,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004891698725701648,
|
|
"loss": 6.7117,
|
|
"mean_token_accuracy": 0.12687891125679016,
|
|
"num_tokens": 4007614.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.914453077316284,
|
|
"epoch": 1.9588894821142553,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004890399120020131,
|
|
"loss": 6.6839,
|
|
"mean_token_accuracy": 0.11670807003974915,
|
|
"num_tokens": 4017645.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.039888191223144,
|
|
"epoch": 1.9642285104111052,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004889091957818631,
|
|
"loss": 6.7942,
|
|
"mean_token_accuracy": 0.1130190908908844,
|
|
"num_tokens": 4029353.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.9136981010437015,
|
|
"epoch": 1.9695675387079552,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004887777243711988,
|
|
"loss": 6.7349,
|
|
"mean_token_accuracy": 0.11921672075986862,
|
|
"num_tokens": 4040421.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.912479543685913,
|
|
"epoch": 1.9749065670048052,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048864549823417,
|
|
"loss": 6.7161,
|
|
"mean_token_accuracy": 0.11533298343420029,
|
|
"num_tokens": 4051858.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.9961567401885985,
|
|
"epoch": 1.9802455953016551,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004885125178375915,
|
|
"loss": 6.7651,
|
|
"mean_token_accuracy": 0.11397730335593223,
|
|
"num_tokens": 4062063.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.809972667694092,
|
|
"epoch": 1.985584623598505,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004883787836509405,
|
|
"loss": 6.5682,
|
|
"mean_token_accuracy": 0.1259746827185154,
|
|
"num_tokens": 4072479.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.965420866012574,
|
|
"epoch": 1.990923651895355,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004882442961463558,
|
|
"loss": 6.7416,
|
|
"mean_token_accuracy": 0.11653700545430183,
|
|
"num_tokens": 4082912.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.961225509643555,
|
|
"epoch": 1.996262680192205,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004881090557986353,
|
|
"loss": 6.7158,
|
|
"mean_token_accuracy": 0.11916097551584244,
|
|
"num_tokens": 4093644.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.774832354651557,
|
|
"epoch": 2.00106780565937,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004879730630852353,
|
|
"loss": 6.6263,
|
|
"mean_token_accuracy": 0.12397908667723338,
|
|
"num_tokens": 4104114.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.998328733444214,
|
|
"epoch": 2.0064068339562198,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048783631848626764,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.12538542225956917,
|
|
"num_tokens": 4114847.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.8019085884094235,
|
|
"epoch": 2.0117458622530697,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004876988224844992,
|
|
"loss": 6.3294,
|
|
"mean_token_accuracy": 0.1320386290550232,
|
|
"num_tokens": 4126078.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.845751762390137,
|
|
"epoch": 2.01708489054992,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048756057556534934,
|
|
"loss": 6.3488,
|
|
"mean_token_accuracy": 0.1348607450723648,
|
|
"num_tokens": 4136678.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.87243618965149,
|
|
"epoch": 2.02242391884677,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004874215782168884,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.13387244567275047,
|
|
"num_tokens": 4146794.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.918681049346924,
|
|
"epoch": 2.02776294714362,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004872818309298365,
|
|
"loss": 6.3912,
|
|
"mean_token_accuracy": 0.14005738869309425,
|
|
"num_tokens": 4156825.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.8306151866912845,
|
|
"epoch": 2.03310197544047,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048714133419756065,
|
|
"loss": 6.3246,
|
|
"mean_token_accuracy": 0.13398497104644774,
|
|
"num_tokens": 4166765.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.886065244674683,
|
|
"epoch": 2.03844100373732,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048700008851607437,
|
|
"loss": 6.4476,
|
|
"mean_token_accuracy": 0.13053907230496406,
|
|
"num_tokens": 4177404.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.970729732513428,
|
|
"epoch": 2.04378003203417,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00048685809438403494,
|
|
"loss": 6.4244,
|
|
"mean_token_accuracy": 0.13034138604998588,
|
|
"num_tokens": 4188804.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.756382846832276,
|
|
"epoch": 2.04911906033102,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867153523027422,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.13649847730994225,
|
|
"num_tokens": 4200009.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.92487530708313,
|
|
"epoch": 2.05445808862787,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004865718627761363,
|
|
"loss": 6.3838,
|
|
"mean_token_accuracy": 0.1362394578754902,
|
|
"num_tokens": 4210776.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.774133539199829,
|
|
"epoch": 2.05979711692472,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048642762631079646,
|
|
"loss": 6.3488,
|
|
"mean_token_accuracy": 0.1345260925590992,
|
|
"num_tokens": 4221412.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.772760629653931,
|
|
"epoch": 2.0651361452215697,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004862826434159388,
|
|
"loss": 6.3442,
|
|
"mean_token_accuracy": 0.13884459361433982,
|
|
"num_tokens": 4231747.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.824880170822143,
|
|
"epoch": 2.0704751735184197,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048613691460341456,
|
|
"loss": 6.3666,
|
|
"mean_token_accuracy": 0.13365612849593161,
|
|
"num_tokens": 4241942.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.788966798782349,
|
|
"epoch": 2.0758142018152697,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048599044038770856,
|
|
"loss": 6.4411,
|
|
"mean_token_accuracy": 0.12992775514721872,
|
|
"num_tokens": 4253151.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.8248374462127686,
|
|
"epoch": 2.0811532301121196,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004858432212859372,
|
|
"loss": 6.3781,
|
|
"mean_token_accuracy": 0.13062778860330582,
|
|
"num_tokens": 4263579.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.798387956619263,
|
|
"epoch": 2.0864922584089696,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004856952578178464,
|
|
"loss": 6.388,
|
|
"mean_token_accuracy": 0.1302419863641262,
|
|
"num_tokens": 4274764.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.881732559204101,
|
|
"epoch": 2.0918312867058195,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004855465505058102,
|
|
"loss": 6.4815,
|
|
"mean_token_accuracy": 0.12663157060742378,
|
|
"num_tokens": 4286561.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.842735052108765,
|
|
"epoch": 2.0971703150026695,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004853970998748289,
|
|
"loss": 6.4403,
|
|
"mean_token_accuracy": 0.13229333907365798,
|
|
"num_tokens": 4296101.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.88682451248169,
|
|
"epoch": 2.1025093432995194,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048524690645252675,
|
|
"loss": 6.3923,
|
|
"mean_token_accuracy": 0.12739012911915779,
|
|
"num_tokens": 4307613.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.759868907928467,
|
|
"epoch": 2.1078483715963694,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004850959707691503,
|
|
"loss": 6.438,
|
|
"mean_token_accuracy": 0.13595941439270973,
|
|
"num_tokens": 4318042.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.849691772460938,
|
|
"epoch": 2.1131873998932194,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000484944293357567,
|
|
"loss": 6.2921,
|
|
"mean_token_accuracy": 0.1407428652048111,
|
|
"num_tokens": 4328690.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.734014511108398,
|
|
"epoch": 2.1185264281900693,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048479187475326254,
|
|
"loss": 6.3446,
|
|
"mean_token_accuracy": 0.13647075593471528,
|
|
"num_tokens": 4339188.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.712657451629639,
|
|
"epoch": 2.1238654564869193,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048463871549433947,
|
|
"loss": 6.3206,
|
|
"mean_token_accuracy": 0.13783530741930008,
|
|
"num_tokens": 4350242.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.698287630081177,
|
|
"epoch": 2.1292044847837692,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048448481612151524,
|
|
"loss": 6.3013,
|
|
"mean_token_accuracy": 0.13802920281887054,
|
|
"num_tokens": 4361138.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.816575288772583,
|
|
"epoch": 2.134543513080619,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048433017717812015,
|
|
"loss": 6.3887,
|
|
"mean_token_accuracy": 0.12946130931377411,
|
|
"num_tokens": 4372157.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.134543513080619,
|
|
"eval_entropy": 6.646139329439634,
|
|
"eval_loss": 6.835247039794922,
|
|
"eval_mean_token_accuracy": 0.1172398333328885,
|
|
"eval_num_tokens": 4372157.0,
|
|
"eval_runtime": 1.8111,
|
|
"eval_samples_per_second": 1700.597,
|
|
"eval_steps_per_second": 212.575,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.71080379486084,
|
|
"epoch": 2.139882541377469,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004841747992100954,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.13796174451708793,
|
|
"num_tokens": 4382735.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.788042116165161,
|
|
"epoch": 2.145221569674319,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048401868276599144,
|
|
"loss": 6.4243,
|
|
"mean_token_accuracy": 0.12602069601416588,
|
|
"num_tokens": 4393839.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.833824014663696,
|
|
"epoch": 2.150560597971169,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048386182839696575,
|
|
"loss": 6.4118,
|
|
"mean_token_accuracy": 0.13158536776900293,
|
|
"num_tokens": 4404612.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.750049638748169,
|
|
"epoch": 2.1558996262680195,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004837042366567809,
|
|
"loss": 6.3932,
|
|
"mean_token_accuracy": 0.13295329585671425,
|
|
"num_tokens": 4415969.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.890046072006226,
|
|
"epoch": 2.161238654564869,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048354590810180306,
|
|
"loss": 6.4796,
|
|
"mean_token_accuracy": 0.1285942628979683,
|
|
"num_tokens": 4427507.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.784240961074829,
|
|
"epoch": 2.1665776828617194,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004833868432909992,
|
|
"loss": 6.416,
|
|
"mean_token_accuracy": 0.1316341355443001,
|
|
"num_tokens": 4438425.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.748818063735962,
|
|
"epoch": 2.1719167111585693,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048322704278593595,
|
|
"loss": 6.3826,
|
|
"mean_token_accuracy": 0.13797276467084885,
|
|
"num_tokens": 4448301.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.770434093475342,
|
|
"epoch": 2.1772557394554193,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048306650715077703,
|
|
"loss": 6.3923,
|
|
"mean_token_accuracy": 0.13586812242865562,
|
|
"num_tokens": 4459214.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.762196683883667,
|
|
"epoch": 2.1825947677522692,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048290523695228163,
|
|
"loss": 6.377,
|
|
"mean_token_accuracy": 0.1391679435968399,
|
|
"num_tokens": 4469696.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.721202230453491,
|
|
"epoch": 2.187933796049119,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004827432327598021,
|
|
"loss": 6.4029,
|
|
"mean_token_accuracy": 0.13381281793117522,
|
|
"num_tokens": 4481020.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.760396432876587,
|
|
"epoch": 2.193272824345969,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004825804951452823,
|
|
"loss": 6.4138,
|
|
"mean_token_accuracy": 0.12782692536711693,
|
|
"num_tokens": 4492877.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.796109580993653,
|
|
"epoch": 2.198611852642819,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004824170246832553,
|
|
"loss": 6.4066,
|
|
"mean_token_accuracy": 0.1343644939363003,
|
|
"num_tokens": 4504184.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.687031936645508,
|
|
"epoch": 2.203950880939669,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 6.3746,
|
|
"mean_token_accuracy": 0.1345256358385086,
|
|
"num_tokens": 4514807.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.734305477142334,
|
|
"epoch": 2.209289909236519,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048208788752774587,
|
|
"loss": 6.3932,
|
|
"mean_token_accuracy": 0.1322820670902729,
|
|
"num_tokens": 4526047.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.770002985000611,
|
|
"epoch": 2.214628937533369,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048192222199625783,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.1383815139532089,
|
|
"num_tokens": 4536506.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.761325740814209,
|
|
"epoch": 2.219967965830219,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004817558259412469,
|
|
"loss": 6.4685,
|
|
"mean_token_accuracy": 0.1286804273724556,
|
|
"num_tokens": 4547351.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.830830669403076,
|
|
"epoch": 2.225306994127069,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004815886999501621,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.12252854704856872,
|
|
"num_tokens": 4558213.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.865453433990479,
|
|
"epoch": 2.230646022423919,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004814208446130292,
|
|
"loss": 6.3481,
|
|
"mean_token_accuracy": 0.1315185956656933,
|
|
"num_tokens": 4567943.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.681244993209839,
|
|
"epoch": 2.235985050720769,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048125226052244904,
|
|
"loss": 6.4045,
|
|
"mean_token_accuracy": 0.13790186494588852,
|
|
"num_tokens": 4580000.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.825203609466553,
|
|
"epoch": 2.2413240790176188,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004810829482735952,
|
|
"loss": 6.4493,
|
|
"mean_token_accuracy": 0.1319129578769207,
|
|
"num_tokens": 4592160.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.706577301025391,
|
|
"epoch": 2.2466631073144687,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004809129084642121,
|
|
"loss": 6.3885,
|
|
"mean_token_accuracy": 0.13410297259688378,
|
|
"num_tokens": 4603747.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.726843214035034,
|
|
"epoch": 2.2520021356113187,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004807421416946125,
|
|
"loss": 6.3895,
|
|
"mean_token_accuracy": 0.13437562063336372,
|
|
"num_tokens": 4615660.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.732323694229126,
|
|
"epoch": 2.2573411639081686,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004805706485676759,
|
|
"loss": 6.4124,
|
|
"mean_token_accuracy": 0.13864737823605539,
|
|
"num_tokens": 4626408.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.802268743515015,
|
|
"epoch": 2.2626801922050186,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004803984296888459,
|
|
"loss": 6.3396,
|
|
"mean_token_accuracy": 0.13721062019467353,
|
|
"num_tokens": 4638236.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.707142543792725,
|
|
"epoch": 2.2680192205018685,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048022548566612855,
|
|
"loss": 6.4276,
|
|
"mean_token_accuracy": 0.13175694569945334,
|
|
"num_tokens": 4649649.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.779877758026123,
|
|
"epoch": 2.2733582487987185,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004800518171100899,
|
|
"loss": 6.4182,
|
|
"mean_token_accuracy": 0.13384050950407983,
|
|
"num_tokens": 4660502.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.758860111236572,
|
|
"epoch": 2.2786972770955685,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004798774246338538,
|
|
"loss": 6.3898,
|
|
"mean_token_accuracy": 0.1331031911075115,
|
|
"num_tokens": 4671486.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.730650806427002,
|
|
"epoch": 2.2840363053924184,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004797023088530999,
|
|
"loss": 6.4384,
|
|
"mean_token_accuracy": 0.1374692849814892,
|
|
"num_tokens": 4681436.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.85043478012085,
|
|
"epoch": 2.2893753336892684,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047952647038606157,
|
|
"loss": 6.4365,
|
|
"mean_token_accuracy": 0.13696684017777444,
|
|
"num_tokens": 4692065.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.76685938835144,
|
|
"epoch": 2.2947143619861183,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004793499098535233,
|
|
"loss": 6.3621,
|
|
"mean_token_accuracy": 0.1343599334359169,
|
|
"num_tokens": 4702107.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.689364004135132,
|
|
"epoch": 2.3000533902829687,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004791726278788191,
|
|
"loss": 6.4656,
|
|
"mean_token_accuracy": 0.13071699887514115,
|
|
"num_tokens": 4712060.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.816775751113892,
|
|
"epoch": 2.3053924185798182,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004789946250878296,
|
|
"loss": 6.3748,
|
|
"mean_token_accuracy": 0.13784249052405356,
|
|
"num_tokens": 4721805.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.65910120010376,
|
|
"epoch": 2.3107314468766686,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004788159021089805,
|
|
"loss": 6.3708,
|
|
"mean_token_accuracy": 0.1317868672311306,
|
|
"num_tokens": 4733630.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.717582035064697,
|
|
"epoch": 2.3160704751735186,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004786364595732401,
|
|
"loss": 6.4233,
|
|
"mean_token_accuracy": 0.13508586436510087,
|
|
"num_tokens": 4743634.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.749582386016845,
|
|
"epoch": 2.3214095034703686,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004784562981141168,
|
|
"loss": 6.3815,
|
|
"mean_token_accuracy": 0.13126638010144234,
|
|
"num_tokens": 4754815.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.702735137939453,
|
|
"epoch": 2.3267485317672185,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004782754183676573,
|
|
"loss": 6.4485,
|
|
"mean_token_accuracy": 0.12935314998030661,
|
|
"num_tokens": 4765657.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.867246341705322,
|
|
"epoch": 2.3320875600640685,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047809382097244407,
|
|
"loss": 6.4413,
|
|
"mean_token_accuracy": 0.131291988492012,
|
|
"num_tokens": 4777436.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.776941537857056,
|
|
"epoch": 2.3374265883609184,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047791150656959326,
|
|
"loss": 6.432,
|
|
"mean_token_accuracy": 0.130543065816164,
|
|
"num_tokens": 4788149.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.67136173248291,
|
|
"epoch": 2.3427656166577684,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047772847580275225,
|
|
"loss": 6.3599,
|
|
"mean_token_accuracy": 0.1306910037994385,
|
|
"num_tokens": 4799512.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.717782402038575,
|
|
"epoch": 2.3481046449546183,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004775447293180976,
|
|
"loss": 6.3866,
|
|
"mean_token_accuracy": 0.1285386636853218,
|
|
"num_tokens": 4810410.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.83135461807251,
|
|
"epoch": 2.3534436732514683,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047736026776433254,
|
|
"loss": 6.5469,
|
|
"mean_token_accuracy": 0.12861532643437384,
|
|
"num_tokens": 4822247.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.84234471321106,
|
|
"epoch": 2.3587827015483183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047717509179268504,
|
|
"loss": 6.3845,
|
|
"mean_token_accuracy": 0.13478757366538047,
|
|
"num_tokens": 4833550.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.656503772735595,
|
|
"epoch": 2.364121729845168,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047698920205690494,
|
|
"loss": 6.4455,
|
|
"mean_token_accuracy": 0.13420262932777405,
|
|
"num_tokens": 4845095.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.730621862411499,
|
|
"epoch": 2.369460758142018,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047680259921326225,
|
|
"loss": 6.3558,
|
|
"mean_token_accuracy": 0.13250561505556108,
|
|
"num_tokens": 4856641.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.756316804885865,
|
|
"epoch": 2.374799786438868,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004766152839205443,
|
|
"loss": 6.3878,
|
|
"mean_token_accuracy": 0.1343679741024971,
|
|
"num_tokens": 4868274.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.710396575927734,
|
|
"epoch": 2.380138814735718,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000476427256840054,
|
|
"loss": 6.394,
|
|
"mean_token_accuracy": 0.13268386349081993,
|
|
"num_tokens": 4878308.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.721910810470581,
|
|
"epoch": 2.385477843032568,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004762385186356068,
|
|
"loss": 6.4499,
|
|
"mean_token_accuracy": 0.13556015565991403,
|
|
"num_tokens": 4888811.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.808997774124146,
|
|
"epoch": 2.390816871329418,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004760490699735289,
|
|
"loss": 6.3471,
|
|
"mean_token_accuracy": 0.1352688789367676,
|
|
"num_tokens": 4898719.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.711210584640503,
|
|
"epoch": 2.396155899626268,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004758589115226549,
|
|
"loss": 6.4057,
|
|
"mean_token_accuracy": 0.13364271447062492,
|
|
"num_tokens": 4909480.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.723228693008423,
|
|
"epoch": 2.401494927923118,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047566804395432493,
|
|
"loss": 6.4533,
|
|
"mean_token_accuracy": 0.12898917347192765,
|
|
"num_tokens": 4920560.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.672474527359009,
|
|
"epoch": 2.406833956219968,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 6.3158,
|
|
"mean_token_accuracy": 0.1333570346236229,
|
|
"num_tokens": 4931499.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.710865020751953,
|
|
"epoch": 2.412172984516818,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047528418416317334,
|
|
"loss": 6.3633,
|
|
"mean_token_accuracy": 0.142713263630867,
|
|
"num_tokens": 4942133.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.741780948638916,
|
|
"epoch": 2.417512012813668,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004750911932955402,
|
|
"loss": 6.4277,
|
|
"mean_token_accuracy": 0.13028594702482224,
|
|
"num_tokens": 4953255.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.699588680267334,
|
|
"epoch": 2.4228510411105177,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047489749602082316,
|
|
"loss": 6.3828,
|
|
"mean_token_accuracy": 0.13321425020694733,
|
|
"num_tokens": 4964766.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.710603332519531,
|
|
"epoch": 2.4281900694073677,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047470309302285617,
|
|
"loss": 6.4475,
|
|
"mean_token_accuracy": 0.1310698799788952,
|
|
"num_tokens": 4976644.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.732292318344117,
|
|
"epoch": 2.4335290977042177,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047450798498796466,
|
|
"loss": 6.3864,
|
|
"mean_token_accuracy": 0.12669613510370253,
|
|
"num_tokens": 4986976.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.759850072860718,
|
|
"epoch": 2.4388681260010676,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004743121726049628,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.13597244247794152,
|
|
"num_tokens": 4998747.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.693120002746582,
|
|
"epoch": 2.444207154297918,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004741156565651518,
|
|
"loss": 6.3692,
|
|
"mean_token_accuracy": 0.13964362144470216,
|
|
"num_tokens": 5009323.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.788399314880371,
|
|
"epoch": 2.4495461825947675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047391843756231693,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.13012440502643585,
|
|
"num_tokens": 5019434.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.664237070083618,
|
|
"epoch": 2.454885210891618,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047372051629272516,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.13567707538604737,
|
|
"num_tokens": 5029884.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.693566989898682,
|
|
"epoch": 2.460224239188468,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047352189345512284,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.13831685185432435,
|
|
"num_tokens": 5040323.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.650610780715942,
|
|
"epoch": 2.465563267485318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004733225697507332,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.13274550884962083,
|
|
"num_tokens": 5051363.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.652336883544922,
|
|
"epoch": 2.470902295782168,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004731225458832537,
|
|
"loss": 6.272,
|
|
"mean_token_accuracy": 0.1453031837940216,
|
|
"num_tokens": 5061221.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.698894119262695,
|
|
"epoch": 2.4762413240790178,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047292182255885367,
|
|
"loss": 6.3481,
|
|
"mean_token_accuracy": 0.138285394012928,
|
|
"num_tokens": 5072578.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.641211462020874,
|
|
"epoch": 2.4815803523758677,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004727204004861721,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.13112420216202736,
|
|
"num_tokens": 5083244.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.702895212173462,
|
|
"epoch": 2.4869193806727177,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004725182803763144,
|
|
"loss": 6.373,
|
|
"mean_token_accuracy": 0.13658028170466424,
|
|
"num_tokens": 5094115.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.758352136611938,
|
|
"epoch": 2.4922584089695676,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004723154629428507,
|
|
"loss": 6.3939,
|
|
"mean_token_accuracy": 0.13706841170787812,
|
|
"num_tokens": 5104838.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.7515819549560545,
|
|
"epoch": 2.4975974372664176,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004721119489018128,
|
|
"loss": 6.3729,
|
|
"mean_token_accuracy": 0.12592497766017913,
|
|
"num_tokens": 5115078.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.657976818084717,
|
|
"epoch": 2.5029364655632675,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047190773897169203,
|
|
"loss": 6.42,
|
|
"mean_token_accuracy": 0.1301332987844944,
|
|
"num_tokens": 5126414.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.748086404800415,
|
|
"epoch": 2.5082754938601175,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047170283387343634,
|
|
"loss": 6.3903,
|
|
"mean_token_accuracy": 0.12998429909348488,
|
|
"num_tokens": 5136271.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.686545467376709,
|
|
"epoch": 2.5136145221569675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047149723433044787,
|
|
"loss": 6.4085,
|
|
"mean_token_accuracy": 0.13363871574401856,
|
|
"num_tokens": 5146780.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.709585857391358,
|
|
"epoch": 2.5189535504538174,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047129094106858056,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.13217394202947616,
|
|
"num_tokens": 5157941.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.742275142669678,
|
|
"epoch": 2.5242925787506674,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047108395481613736,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.1324183464050293,
|
|
"num_tokens": 5168994.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.677269458770752,
|
|
"epoch": 2.5296316070475173,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047087627630386803,
|
|
"loss": 6.3419,
|
|
"mean_token_accuracy": 0.12768020182847978,
|
|
"num_tokens": 5180780.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.604436016082763,
|
|
"epoch": 2.5349706353443673,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000470667906264966,
|
|
"loss": 6.4301,
|
|
"mean_token_accuracy": 0.13026676774024964,
|
|
"num_tokens": 5192090.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.700447416305542,
|
|
"epoch": 2.5403096636412172,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004704588454350661,
|
|
"loss": 6.3294,
|
|
"mean_token_accuracy": 0.1367558591067791,
|
|
"num_tokens": 5201854.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.711759424209594,
|
|
"epoch": 2.545648691938067,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004702490945522423,
|
|
"loss": 6.3489,
|
|
"mean_token_accuracy": 0.1381134457886219,
|
|
"num_tokens": 5213998.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.620312118530274,
|
|
"epoch": 2.550987720234917,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004700386543570043,
|
|
"loss": 6.3038,
|
|
"mean_token_accuracy": 0.13607224375009536,
|
|
"num_tokens": 5225217.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.5841038703918455,
|
|
"epoch": 2.556326748531767,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004698275255922956,
|
|
"loss": 6.245,
|
|
"mean_token_accuracy": 0.1424443930387497,
|
|
"num_tokens": 5235098.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.684883546829224,
|
|
"epoch": 2.561665776828617,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046961570900349073,
|
|
"loss": 6.4372,
|
|
"mean_token_accuracy": 0.13414805009961128,
|
|
"num_tokens": 5246704.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.6152567863464355,
|
|
"epoch": 2.567004805125467,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004694032053383922,
|
|
"loss": 6.3382,
|
|
"mean_token_accuracy": 0.136787299066782,
|
|
"num_tokens": 5257331.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.684454822540284,
|
|
"epoch": 2.572343833422317,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004691900153472286,
|
|
"loss": 6.3202,
|
|
"mean_token_accuracy": 0.13671067953109742,
|
|
"num_tokens": 5267662.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.733580303192139,
|
|
"epoch": 2.5776828617191674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046897613978265124,
|
|
"loss": 6.3344,
|
|
"mean_token_accuracy": 0.13251007497310638,
|
|
"num_tokens": 5277977.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.702429294586182,
|
|
"epoch": 2.583021890016017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004687615793997319,
|
|
"loss": 6.4651,
|
|
"mean_token_accuracy": 0.12662636265158653,
|
|
"num_tokens": 5288987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.692728757858276,
|
|
"epoch": 2.5883609183128673,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004685463349559602,
|
|
"loss": 6.2996,
|
|
"mean_token_accuracy": 0.1411239504814148,
|
|
"num_tokens": 5300300.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.668736267089844,
|
|
"epoch": 2.593699946609717,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046833040721124035,
|
|
"loss": 6.3061,
|
|
"mean_token_accuracy": 0.13349644243717193,
|
|
"num_tokens": 5311606.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.6499671459198,
|
|
"epoch": 2.599038974906567,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004681137969278896,
|
|
"loss": 6.4611,
|
|
"mean_token_accuracy": 0.12544659078121184,
|
|
"num_tokens": 5323218.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.645833206176758,
|
|
"epoch": 2.6043780032034167,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004678965048706341,
|
|
"loss": 6.3175,
|
|
"mean_token_accuracy": 0.13822234496474267,
|
|
"num_tokens": 5334516.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.649818849563599,
|
|
"epoch": 2.609717031500267,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 6.3781,
|
|
"mean_token_accuracy": 0.133113045245409,
|
|
"num_tokens": 5345751.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.690796184539795,
|
|
"epoch": 2.615056059797117,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004674598785053475,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.13145751953125,
|
|
"num_tokens": 5356918.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.715551376342773,
|
|
"epoch": 2.620395088093967,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046724054573879314,
|
|
"loss": 6.3541,
|
|
"mean_token_accuracy": 0.13839275687932967,
|
|
"num_tokens": 5368175.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.555882930755615,
|
|
"epoch": 2.625734116390817,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004670205342812827,
|
|
"loss": 6.3243,
|
|
"mean_token_accuracy": 0.13807284832000732,
|
|
"num_tokens": 5379074.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.698004627227784,
|
|
"epoch": 2.631073144687667,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046679984490955015,
|
|
"loss": 6.3792,
|
|
"mean_token_accuracy": 0.13963908702135086,
|
|
"num_tokens": 5389380.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.603578090667725,
|
|
"epoch": 2.636412172984517,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004665784784027229,
|
|
"loss": 6.4132,
|
|
"mean_token_accuracy": 0.13925916329026222,
|
|
"num_tokens": 5401378.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.723094749450683,
|
|
"epoch": 2.641751201281367,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004663564355423189,
|
|
"loss": 6.3635,
|
|
"mean_token_accuracy": 0.13477470874786376,
|
|
"num_tokens": 5411434.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.678373670578003,
|
|
"epoch": 2.647090229578217,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004661337171122442,
|
|
"loss": 6.3339,
|
|
"mean_token_accuracy": 0.13658420145511627,
|
|
"num_tokens": 5421597.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.576092147827149,
|
|
"epoch": 2.652429257875067,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046591032389878934,
|
|
"loss": 6.3451,
|
|
"mean_token_accuracy": 0.13853081464767455,
|
|
"num_tokens": 5432166.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.631149673461914,
|
|
"epoch": 2.6577682861719167,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004656862566906276,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.13675326853990555,
|
|
"num_tokens": 5443156.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.7582684516906735,
|
|
"epoch": 2.6631073144687667,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046546151627881166,
|
|
"loss": 6.4168,
|
|
"mean_token_accuracy": 0.12362042963504791,
|
|
"num_tokens": 5453830.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.597014427185059,
|
|
"epoch": 2.6684463427656167,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004652361034567708,
|
|
"loss": 6.3313,
|
|
"mean_token_accuracy": 0.13709120973944663,
|
|
"num_tokens": 5465371.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.6684463427656167,
|
|
"eval_entropy": 6.396776999436415,
|
|
"eval_loss": 6.722992420196533,
|
|
"eval_mean_token_accuracy": 0.12398916864162916,
|
|
"eval_num_tokens": 5465371.0,
|
|
"eval_runtime": 1.8079,
|
|
"eval_samples_per_second": 1703.658,
|
|
"eval_steps_per_second": 212.957,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.674832677841186,
|
|
"epoch": 2.6737853710624666,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046501001902030814,
|
|
"loss": 6.3979,
|
|
"mean_token_accuracy": 0.12944845110177994,
|
|
"num_tokens": 5476687.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.702767705917358,
|
|
"epoch": 2.6791243993593166,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004647832637675982,
|
|
"loss": 6.3619,
|
|
"mean_token_accuracy": 0.12942193895578386,
|
|
"num_tokens": 5487825.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.628402948379517,
|
|
"epoch": 2.6844634276561665,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046455583849918324,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.13754471391439438,
|
|
"num_tokens": 5498561.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.661627006530762,
|
|
"epoch": 2.6898024559530165,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004643277440179715,
|
|
"loss": 6.3473,
|
|
"mean_token_accuracy": 0.13480948209762572,
|
|
"num_tokens": 5509624.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.622825479507446,
|
|
"epoch": 2.6951414842498664,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004640989811292335,
|
|
"loss": 6.2646,
|
|
"mean_token_accuracy": 0.14940746873617172,
|
|
"num_tokens": 5519484.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.572876882553101,
|
|
"epoch": 2.7004805125467164,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004638695506405996,
|
|
"loss": 6.277,
|
|
"mean_token_accuracy": 0.13872266560792923,
|
|
"num_tokens": 5531839.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.6564329147338865,
|
|
"epoch": 2.7058195408435664,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004636394533620571,
|
|
"loss": 6.3811,
|
|
"mean_token_accuracy": 0.13545397743582727,
|
|
"num_tokens": 5541787.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.690985822677613,
|
|
"epoch": 2.7111585691404163,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004634086901059474,
|
|
"loss": 6.4281,
|
|
"mean_token_accuracy": 0.1288560926914215,
|
|
"num_tokens": 5553493.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.66993899345398,
|
|
"epoch": 2.7164975974372663,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004631772616869629,
|
|
"loss": 6.313,
|
|
"mean_token_accuracy": 0.13453716188669204,
|
|
"num_tokens": 5564809.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.551520204544067,
|
|
"epoch": 2.7218366257341167,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046294516892214454,
|
|
"loss": 6.3144,
|
|
"mean_token_accuracy": 0.1358671985566616,
|
|
"num_tokens": 5575575.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.64482274055481,
|
|
"epoch": 2.727175654030966,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046271241263087846,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.1315604753792286,
|
|
"num_tokens": 5587276.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.6441398620605465,
|
|
"epoch": 2.7325146823278166,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004624789936348936,
|
|
"loss": 6.3047,
|
|
"mean_token_accuracy": 0.13094909787178038,
|
|
"num_tokens": 5597557.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.540618276596069,
|
|
"epoch": 2.737853710624666,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046224491275825825,
|
|
"loss": 6.3191,
|
|
"mean_token_accuracy": 0.13919277861714363,
|
|
"num_tokens": 5608512.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.607910633087158,
|
|
"epoch": 2.7431927389215165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004620101708273776,
|
|
"loss": 6.3718,
|
|
"mean_token_accuracy": 0.13771670833230018,
|
|
"num_tokens": 5621788.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.716967058181763,
|
|
"epoch": 2.748531767218366,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046177476867099055,
|
|
"loss": 6.353,
|
|
"mean_token_accuracy": 0.1359571024775505,
|
|
"num_tokens": 5632493.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.66322660446167,
|
|
"epoch": 2.7538707955152164,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046153870712016694,
|
|
"loss": 6.3838,
|
|
"mean_token_accuracy": 0.12872357442975044,
|
|
"num_tokens": 5643163.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.628427648544312,
|
|
"epoch": 2.7592098238120664,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004613019870083045,
|
|
"loss": 6.2877,
|
|
"mean_token_accuracy": 0.1398894064128399,
|
|
"num_tokens": 5653362.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.630182266235352,
|
|
"epoch": 2.7645488521089163,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004610646091711261,
|
|
"loss": 6.4109,
|
|
"mean_token_accuracy": 0.1394827112555504,
|
|
"num_tokens": 5663637.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.614586591720581,
|
|
"epoch": 2.7698878804057663,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004608265744466763,
|
|
"loss": 6.3426,
|
|
"mean_token_accuracy": 0.12996145933866501,
|
|
"num_tokens": 5674446.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.644121503829956,
|
|
"epoch": 2.7752269087026162,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046058788367531927,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.1369388572871685,
|
|
"num_tokens": 5684768.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.684825897216797,
|
|
"epoch": 2.780565936999466,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004603485376997349,
|
|
"loss": 6.2749,
|
|
"mean_token_accuracy": 0.14011628329753875,
|
|
"num_tokens": 5696474.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.564452171325684,
|
|
"epoch": 2.785904965296316,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004601085373649163,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.12873738557100295,
|
|
"num_tokens": 5707829.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.675048208236694,
|
|
"epoch": 2.791243993593166,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000459867883518167,
|
|
"loss": 6.3696,
|
|
"mean_token_accuracy": 0.13425890281796454,
|
|
"num_tokens": 5719325.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.638251638412475,
|
|
"epoch": 2.796583021890016,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004596265770090974,
|
|
"loss": 6.3418,
|
|
"mean_token_accuracy": 0.1312248796224594,
|
|
"num_tokens": 5730859.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.603208208084107,
|
|
"epoch": 2.801922050186866,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004593846186896222,
|
|
"loss": 6.3033,
|
|
"mean_token_accuracy": 0.1391927905380726,
|
|
"num_tokens": 5741595.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.703442621231079,
|
|
"epoch": 2.807261078483716,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045914200941395725,
|
|
"loss": 6.3379,
|
|
"mean_token_accuracy": 0.13446537926793098,
|
|
"num_tokens": 5752989.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.583294916152954,
|
|
"epoch": 2.812600106780566,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004588987500386165,
|
|
"loss": 6.3326,
|
|
"mean_token_accuracy": 0.13479498475790025,
|
|
"num_tokens": 5764440.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.604435062408447,
|
|
"epoch": 2.817939135077416,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004586548414224093,
|
|
"loss": 6.2446,
|
|
"mean_token_accuracy": 0.13860116750001908,
|
|
"num_tokens": 5776389.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.605298662185669,
|
|
"epoch": 2.823278163374266,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045841028442643675,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.13552825450897216,
|
|
"num_tokens": 5787803.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.604708576202393,
|
|
"epoch": 2.828617191671116,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004581650799140891,
|
|
"loss": 6.3627,
|
|
"mean_token_accuracy": 0.1368337392807007,
|
|
"num_tokens": 5799647.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.634318542480469,
|
|
"epoch": 2.8339562199679658,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004579192287510427,
|
|
"loss": 6.3587,
|
|
"mean_token_accuracy": 0.13644306510686874,
|
|
"num_tokens": 5809968.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.655166101455689,
|
|
"epoch": 2.8392952482648157,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004576727318052568,
|
|
"loss": 6.3382,
|
|
"mean_token_accuracy": 0.14500015899538993,
|
|
"num_tokens": 5820075.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.592358732223511,
|
|
"epoch": 2.8446342765616657,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045742558994697063,
|
|
"loss": 6.3571,
|
|
"mean_token_accuracy": 0.13405183851718902,
|
|
"num_tokens": 5830686.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.652147436141968,
|
|
"epoch": 2.8499733048585156,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004571778040487001,
|
|
"loss": 6.3224,
|
|
"mean_token_accuracy": 0.13853413611650467,
|
|
"num_tokens": 5841451.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.5636547088623045,
|
|
"epoch": 2.8553123331553656,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004569293749852347,
|
|
"loss": 6.3725,
|
|
"mean_token_accuracy": 0.1358483128249645,
|
|
"num_tokens": 5852650.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.648437976837158,
|
|
"epoch": 2.8606513614522155,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004566803036336351,
|
|
"loss": 6.3285,
|
|
"mean_token_accuracy": 0.13660418763756751,
|
|
"num_tokens": 5865056.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.6106782913208,
|
|
"epoch": 2.865990389749066,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045643059087322895,
|
|
"loss": 6.3662,
|
|
"mean_token_accuracy": 0.13459570184350014,
|
|
"num_tokens": 5876069.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.6013682842254635,
|
|
"epoch": 2.8713294180459155,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004561802375856088,
|
|
"loss": 6.3081,
|
|
"mean_token_accuracy": 0.14166605323553086,
|
|
"num_tokens": 5886366.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.585530471801758,
|
|
"epoch": 2.876668446342766,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004559292446546281,
|
|
"loss": 6.3872,
|
|
"mean_token_accuracy": 0.13289673924446105,
|
|
"num_tokens": 5897101.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.7206700325012205,
|
|
"epoch": 2.8820074746396154,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045567761296639887,
|
|
"loss": 6.3865,
|
|
"mean_token_accuracy": 0.13457530811429025,
|
|
"num_tokens": 5908748.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.716259765625,
|
|
"epoch": 2.8873465029364658,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000455425343409288,
|
|
"loss": 6.4175,
|
|
"mean_token_accuracy": 0.13208478912711144,
|
|
"num_tokens": 5920830.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.493069410324097,
|
|
"epoch": 2.8926855312333153,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045517243687391454,
|
|
"loss": 6.224,
|
|
"mean_token_accuracy": 0.14297640398144723,
|
|
"num_tokens": 5932360.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.642335271835327,
|
|
"epoch": 2.8980245595301657,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004549188942531461,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.13232816904783248,
|
|
"num_tokens": 5944067.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.675318431854248,
|
|
"epoch": 2.9033635878270156,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000454664716442096,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.13153403401374816,
|
|
"num_tokens": 5955974.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.689729070663452,
|
|
"epoch": 2.9087026161238656,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045440990433812025,
|
|
"loss": 6.4076,
|
|
"mean_token_accuracy": 0.13313242122530938,
|
|
"num_tokens": 5966542.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.578455638885498,
|
|
"epoch": 2.9140416444207156,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000454154458840814,
|
|
"loss": 6.2859,
|
|
"mean_token_accuracy": 0.13757668510079385,
|
|
"num_tokens": 5977342.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.657413530349731,
|
|
"epoch": 2.9193806727175655,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004538983808520085,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.13169204592704772,
|
|
"num_tokens": 5988786.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.686155366897583,
|
|
"epoch": 2.9247197010144155,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045364167127576817,
|
|
"loss": 6.423,
|
|
"mean_token_accuracy": 0.13093676418066025,
|
|
"num_tokens": 6000167.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.569418096542359,
|
|
"epoch": 2.9300587293112654,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045338433101838705,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.13144461959600448,
|
|
"num_tokens": 6012214.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.686597728729248,
|
|
"epoch": 2.9353977576081154,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045312636098838576,
|
|
"loss": 6.4008,
|
|
"mean_token_accuracy": 0.1341310903429985,
|
|
"num_tokens": 6023032.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.583488416671753,
|
|
"epoch": 2.9407367859049653,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004528677620965083,
|
|
"loss": 6.2965,
|
|
"mean_token_accuracy": 0.13848185539245605,
|
|
"num_tokens": 6033398.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.651776885986328,
|
|
"epoch": 2.9460758142018153,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045260853525571894,
|
|
"loss": 6.334,
|
|
"mean_token_accuracy": 0.13625661581754683,
|
|
"num_tokens": 6044549.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.496178436279297,
|
|
"epoch": 2.9514148424986653,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004523486813811987,
|
|
"loss": 6.1268,
|
|
"mean_token_accuracy": 0.14498887583613396,
|
|
"num_tokens": 6055213.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.558784627914429,
|
|
"epoch": 2.956753870795515,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004520882013903423,
|
|
"loss": 6.2762,
|
|
"mean_token_accuracy": 0.1483754262328148,
|
|
"num_tokens": 6065488.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.574901962280274,
|
|
"epoch": 2.962092899092365,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045182709620275504,
|
|
"loss": 6.3614,
|
|
"mean_token_accuracy": 0.14127218276262282,
|
|
"num_tokens": 6077525.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.635270023345948,
|
|
"epoch": 2.967431927389215,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004515653667402494,
|
|
"loss": 6.3527,
|
|
"mean_token_accuracy": 0.13435368686914445,
|
|
"num_tokens": 6088707.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.591364526748658,
|
|
"epoch": 2.972770955686065,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004513030139268418,
|
|
"loss": 6.3043,
|
|
"mean_token_accuracy": 0.1398948423564434,
|
|
"num_tokens": 6099231.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.56983323097229,
|
|
"epoch": 2.978109983982915,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004510400386887493,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.1362353518605232,
|
|
"num_tokens": 6110356.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.708747625350952,
|
|
"epoch": 2.983449012279765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045077644195438645,
|
|
"loss": 6.3319,
|
|
"mean_token_accuracy": 0.13777759671211243,
|
|
"num_tokens": 6120722.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.617311716079712,
|
|
"epoch": 2.988788040576615,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000450512224654362,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.13757702708244324,
|
|
"num_tokens": 6131404.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.617153215408325,
|
|
"epoch": 2.994127068873465,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045024738772147534,
|
|
"loss": 6.3327,
|
|
"mean_token_accuracy": 0.12845131382346153,
|
|
"num_tokens": 6141402.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.617161083221435,
|
|
"epoch": 2.999466097170315,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044998193209071365,
|
|
"loss": 6.2894,
|
|
"mean_token_accuracy": 0.13450669795274733,
|
|
"num_tokens": 6152082.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.575082884894477,
|
|
"epoch": 3.00427122263748,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044971585869924826,
|
|
"loss": 5.9833,
|
|
"mean_token_accuracy": 0.15619772341516283,
|
|
"num_tokens": 6162282.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.512451982498169,
|
|
"epoch": 3.00961025093433,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004494491684864316,
|
|
"loss": 5.9614,
|
|
"mean_token_accuracy": 0.15196620151400567,
|
|
"num_tokens": 6173818.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.600033617019653,
|
|
"epoch": 3.01494927923118,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.15490610152482986,
|
|
"num_tokens": 6184406.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.595245456695556,
|
|
"epoch": 3.02028830752803,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004489139413650385,
|
|
"loss": 6.0396,
|
|
"mean_token_accuracy": 0.14942893981933594,
|
|
"num_tokens": 6195817.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.5526063442230225,
|
|
"epoch": 3.02562733582488,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004486454063460415,
|
|
"loss": 5.9056,
|
|
"mean_token_accuracy": 0.1643466129899025,
|
|
"num_tokens": 6206176.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.5352250099182125,
|
|
"epoch": 3.03096636412173,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004483762582848457,
|
|
"loss": 5.9864,
|
|
"mean_token_accuracy": 0.1512788400053978,
|
|
"num_tokens": 6215783.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.566704607009887,
|
|
"epoch": 3.03630539241858,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004481064981316583,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.15059448778629303,
|
|
"num_tokens": 6227146.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.573222684860229,
|
|
"epoch": 3.04164442071543,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044783612683884744,
|
|
"loss": 5.9499,
|
|
"mean_token_accuracy": 0.1592599168419838,
|
|
"num_tokens": 6237417.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.531847047805786,
|
|
"epoch": 3.04698344901228,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004475651453609388,
|
|
"loss": 6.005,
|
|
"mean_token_accuracy": 0.15580991804599761,
|
|
"num_tokens": 6249136.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.524133777618408,
|
|
"epoch": 3.0523224773091298,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044729355465461236,
|
|
"loss": 5.9748,
|
|
"mean_token_accuracy": 0.14894505292177201,
|
|
"num_tokens": 6260811.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.56110110282898,
|
|
"epoch": 3.0576615056059797,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000447021355678699,
|
|
"loss": 6.0341,
|
|
"mean_token_accuracy": 0.1491138905286789,
|
|
"num_tokens": 6272776.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.581620693206787,
|
|
"epoch": 3.0630005339028297,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004467485493941769,
|
|
"loss": 5.9793,
|
|
"mean_token_accuracy": 0.15370213985443115,
|
|
"num_tokens": 6283335.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.473287630081177,
|
|
"epoch": 3.0683395621996796,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044647513676416835,
|
|
"loss": 5.9194,
|
|
"mean_token_accuracy": 0.1597832977771759,
|
|
"num_tokens": 6293383.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.5341156959533695,
|
|
"epoch": 3.0736785904965296,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004462011187539364,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.16253844350576402,
|
|
"num_tokens": 6303373.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.571972370147705,
|
|
"epoch": 3.0790176187933795,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004459264963308812,
|
|
"loss": 5.9775,
|
|
"mean_token_accuracy": 0.15417150035500526,
|
|
"num_tokens": 6315674.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.518611907958984,
|
|
"epoch": 3.0843566470902295,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044565127046453687,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.15624917298555374,
|
|
"num_tokens": 6326183.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.38936939239502,
|
|
"epoch": 3.0896956753870795,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044537544212656775,
|
|
"loss": 5.987,
|
|
"mean_token_accuracy": 0.15313675105571747,
|
|
"num_tokens": 6337495.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.50475435256958,
|
|
"epoch": 3.0950347036839294,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044509901229076537,
|
|
"loss": 5.9536,
|
|
"mean_token_accuracy": 0.15808096826076506,
|
|
"num_tokens": 6348935.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.509057188034058,
|
|
"epoch": 3.1003737319807794,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004448219819330447,
|
|
"loss": 6.0969,
|
|
"mean_token_accuracy": 0.14443548917770385,
|
|
"num_tokens": 6359701.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.627410411834717,
|
|
"epoch": 3.1057127602776293,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445443520314409,
|
|
"loss": 6.0152,
|
|
"mean_token_accuracy": 0.15080558508634567,
|
|
"num_tokens": 6370134.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.48399600982666,
|
|
"epoch": 3.1110517885744793,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044426612356610555,
|
|
"loss": 6.054,
|
|
"mean_token_accuracy": 0.15006683990359307,
|
|
"num_tokens": 6381644.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.544503784179687,
|
|
"epoch": 3.1163908168713292,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004439872975193037,
|
|
"loss": 6.0202,
|
|
"mean_token_accuracy": 0.14872004687786103,
|
|
"num_tokens": 6393386.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.50056939125061,
|
|
"epoch": 3.121729845168179,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044370787487540986,
|
|
"loss": 6.0176,
|
|
"mean_token_accuracy": 0.14920699894428252,
|
|
"num_tokens": 6405487.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.4862977981567385,
|
|
"epoch": 3.1270688734650296,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000443427856620905,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.14947659522294998,
|
|
"num_tokens": 6416470.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.616715431213379,
|
|
"epoch": 3.132407901761879,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044314724374437274,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.1493872210383415,
|
|
"num_tokens": 6428579.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.536284017562866,
|
|
"epoch": 3.1377469300587295,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004428660372364959,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.1518949344754219,
|
|
"num_tokens": 6438976.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.524206018447876,
|
|
"epoch": 3.1430859583555795,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004425842380900532,
|
|
"loss": 6.0069,
|
|
"mean_token_accuracy": 0.14818792790174484,
|
|
"num_tokens": 6449919.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.430868816375733,
|
|
"epoch": 3.1484249866524294,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044230184729991565,
|
|
"loss": 5.987,
|
|
"mean_token_accuracy": 0.15109720230102539,
|
|
"num_tokens": 6460874.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.580015373229981,
|
|
"epoch": 3.1537640149492794,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004420188658630428,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.15066762641072273,
|
|
"num_tokens": 6471606.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.501916456222534,
|
|
"epoch": 3.1591030432461293,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044173529477847974,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.15116074085235595,
|
|
"num_tokens": 6481917.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.472680139541626,
|
|
"epoch": 3.1644420715429793,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044145113504735295,
|
|
"loss": 5.997,
|
|
"mean_token_accuracy": 0.15924959033727645,
|
|
"num_tokens": 6493295.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.533931303024292,
|
|
"epoch": 3.1697810998398293,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004411663876728674,
|
|
"loss": 5.9968,
|
|
"mean_token_accuracy": 0.15576179772615434,
|
|
"num_tokens": 6504208.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.4077013492584225,
|
|
"epoch": 3.175120128136679,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044088105366030234,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.15747272372245788,
|
|
"num_tokens": 6515347.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.491119623184204,
|
|
"epoch": 3.180459156433529,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004405951340170084,
|
|
"loss": 6.0156,
|
|
"mean_token_accuracy": 0.1451236553490162,
|
|
"num_tokens": 6527674.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.439837503433227,
|
|
"epoch": 3.185798184730379,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044030862975240364,
|
|
"loss": 5.9643,
|
|
"mean_token_accuracy": 0.15301360040903092,
|
|
"num_tokens": 6539048.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.512312126159668,
|
|
"epoch": 3.191137213027229,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044002154187797003,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.1505460411310196,
|
|
"num_tokens": 6550152.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.596912431716919,
|
|
"epoch": 3.196476241324079,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043973387140725,
|
|
"loss": 6.0095,
|
|
"mean_token_accuracy": 0.153944893181324,
|
|
"num_tokens": 6561189.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.5509546279907225,
|
|
"epoch": 3.201815269620929,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004394456193558428,
|
|
"loss": 6.0695,
|
|
"mean_token_accuracy": 0.14542510956525803,
|
|
"num_tokens": 6572548.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.201815269620929,
|
|
"eval_entropy": 6.326781872340611,
|
|
"eval_loss": 6.670738220214844,
|
|
"eval_mean_token_accuracy": 0.12738430138145174,
|
|
"eval_num_tokens": 6572548.0,
|
|
"eval_runtime": 1.8101,
|
|
"eval_samples_per_second": 1701.536,
|
|
"eval_steps_per_second": 212.692,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.55753607749939,
|
|
"epoch": 3.207154297917779,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004391567867414007,
|
|
"loss": 6.0776,
|
|
"mean_token_accuracy": 0.1507057286798954,
|
|
"num_tokens": 6584176.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.498582029342652,
|
|
"epoch": 3.212493326214629,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043886737458362596,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.15997665971517563,
|
|
"num_tokens": 6595070.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.430584669113159,
|
|
"epoch": 3.217832354511479,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004385773839042666,
|
|
"loss": 5.961,
|
|
"mean_token_accuracy": 0.14958512634038926,
|
|
"num_tokens": 6606556.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.498847532272339,
|
|
"epoch": 3.223171382808329,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043828681572711316,
|
|
"loss": 5.9217,
|
|
"mean_token_accuracy": 0.1578720450401306,
|
|
"num_tokens": 6616944.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.42176513671875,
|
|
"epoch": 3.228510411105179,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043799567107799504,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.15884335339069366,
|
|
"num_tokens": 6627541.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.491019201278687,
|
|
"epoch": 3.2338494394020287,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043770395098477666,
|
|
"loss": 5.9677,
|
|
"mean_token_accuracy": 0.15469538271427155,
|
|
"num_tokens": 6639102.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.58637809753418,
|
|
"epoch": 3.2391884676988787,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004374116564773543,
|
|
"loss": 6.0057,
|
|
"mean_token_accuracy": 0.15035879462957383,
|
|
"num_tokens": 6650737.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.4510668277740475,
|
|
"epoch": 3.2445274959957286,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004371187885876519,
|
|
"loss": 6.0289,
|
|
"mean_token_accuracy": 0.14777518883347512,
|
|
"num_tokens": 6661812.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.511271858215332,
|
|
"epoch": 3.2498665242925786,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043682534834961766,
|
|
"loss": 6.1004,
|
|
"mean_token_accuracy": 0.15231523364782334,
|
|
"num_tokens": 6672122.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.541240692138672,
|
|
"epoch": 3.2552055525894286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043653133679922077,
|
|
"loss": 6.0743,
|
|
"mean_token_accuracy": 0.14883374273777009,
|
|
"num_tokens": 6684219.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.464490556716919,
|
|
"epoch": 3.2605445808862785,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043623675497444695,
|
|
"loss": 5.8956,
|
|
"mean_token_accuracy": 0.1576215833425522,
|
|
"num_tokens": 6694290.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.485196208953857,
|
|
"epoch": 3.2658836091831285,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043594160391529545,
|
|
"loss": 6.0103,
|
|
"mean_token_accuracy": 0.15430199205875397,
|
|
"num_tokens": 6704313.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.491442728042602,
|
|
"epoch": 3.271222637479979,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043564588466377536,
|
|
"loss": 6.0135,
|
|
"mean_token_accuracy": 0.15234167277812957,
|
|
"num_tokens": 6715424.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.400201511383057,
|
|
"epoch": 3.2765616657768284,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004353495982639012,
|
|
"loss": 5.902,
|
|
"mean_token_accuracy": 0.15296155214309692,
|
|
"num_tokens": 6727025.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.462273597717285,
|
|
"epoch": 3.281900694073679,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043505274576169037,
|
|
"loss": 6.0069,
|
|
"mean_token_accuracy": 0.15597002059221268,
|
|
"num_tokens": 6738344.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.481720447540283,
|
|
"epoch": 3.2872397223705287,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043475532820515836,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.1552176684141159,
|
|
"num_tokens": 6748812.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.529934406280518,
|
|
"epoch": 3.2925787506673787,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043445734664431575,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.1514669880270958,
|
|
"num_tokens": 6759776.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.4437689781188965,
|
|
"epoch": 3.2979177789642287,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004341588021311644,
|
|
"loss": 5.9939,
|
|
"mean_token_accuracy": 0.156333290040493,
|
|
"num_tokens": 6770702.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.403773069381714,
|
|
"epoch": 3.3032568072610786,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043385969571969353,
|
|
"loss": 5.9915,
|
|
"mean_token_accuracy": 0.15836867243051528,
|
|
"num_tokens": 6781599.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.451881837844849,
|
|
"epoch": 3.3085958355579286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043356002846587597,
|
|
"loss": 6.0281,
|
|
"mean_token_accuracy": 0.15876768231391908,
|
|
"num_tokens": 6792385.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.551677274703979,
|
|
"epoch": 3.3139348638547785,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004332598014276647,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.14841348230838775,
|
|
"num_tokens": 6803805.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.390894222259521,
|
|
"epoch": 3.3192738921516285,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043295901566498907,
|
|
"loss": 6.0232,
|
|
"mean_token_accuracy": 0.15243831574916838,
|
|
"num_tokens": 6815460.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.549357843399048,
|
|
"epoch": 3.3246129204484784,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043265767223975076,
|
|
"loss": 5.9665,
|
|
"mean_token_accuracy": 0.1580589607357979,
|
|
"num_tokens": 6825995.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.414888572692871,
|
|
"epoch": 3.3299519487453284,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004323557722158204,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.1552261747419834,
|
|
"num_tokens": 6836793.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.447903203964233,
|
|
"epoch": 3.3352909770421784,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004320533166590335,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.15631503015756607,
|
|
"num_tokens": 6849063.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.543558025360108,
|
|
"epoch": 3.3406300053390283,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000431750306637187,
|
|
"loss": 5.996,
|
|
"mean_token_accuracy": 0.1529462069272995,
|
|
"num_tokens": 6859821.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.450773859024048,
|
|
"epoch": 3.3459690336358783,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043144674322003525,
|
|
"loss": 6.0202,
|
|
"mean_token_accuracy": 0.15169835835695267,
|
|
"num_tokens": 6871026.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.499336385726929,
|
|
"epoch": 3.3513080619327282,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004311426274792863,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.15381820052862166,
|
|
"num_tokens": 6883135.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.461349821090698,
|
|
"epoch": 3.356647090229578,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004308379604885981,
|
|
"loss": 6.0299,
|
|
"mean_token_accuracy": 0.14820491522550583,
|
|
"num_tokens": 6893943.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.399202156066894,
|
|
"epoch": 3.361986118526428,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004305327433235749,
|
|
"loss": 6.0323,
|
|
"mean_token_accuracy": 0.149459370970726,
|
|
"num_tokens": 6904172.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.516909122467041,
|
|
"epoch": 3.367325146823278,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004302269770617632,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.15370023101568223,
|
|
"num_tokens": 6915440.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.542649269104004,
|
|
"epoch": 3.372664175120128,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000429920662782648,
|
|
"loss": 6.0811,
|
|
"mean_token_accuracy": 0.14919362887740134,
|
|
"num_tokens": 6927762.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.5270734310150145,
|
|
"epoch": 3.378003203416978,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042961380156764925,
|
|
"loss": 6.0181,
|
|
"mean_token_accuracy": 0.14989374130964278,
|
|
"num_tokens": 6938139.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.439842700958252,
|
|
"epoch": 3.383342231713828,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004293063945001175,
|
|
"loss": 6.0218,
|
|
"mean_token_accuracy": 0.15330789387226104,
|
|
"num_tokens": 6949463.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.543133401870728,
|
|
"epoch": 3.388681260010678,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042899844266533054,
|
|
"loss": 6.0749,
|
|
"mean_token_accuracy": 0.14984160363674165,
|
|
"num_tokens": 6960660.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.506657409667969,
|
|
"epoch": 3.394020288307528,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004286899471504896,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.1571991503238678,
|
|
"num_tokens": 6971283.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.540865230560303,
|
|
"epoch": 3.399359316604378,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042838090904471495,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.15327067971229552,
|
|
"num_tokens": 6981248.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.339823627471924,
|
|
"epoch": 3.404698344901228,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004280713294390428,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.15587875843048096,
|
|
"num_tokens": 6991574.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.46599383354187,
|
|
"epoch": 3.4100373731980778,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004277612094264209,
|
|
"loss": 6.0057,
|
|
"mean_token_accuracy": 0.15858537703752518,
|
|
"num_tokens": 7002548.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.520251131057739,
|
|
"epoch": 3.415376401494928,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004274505501017048,
|
|
"loss": 6.0499,
|
|
"mean_token_accuracy": 0.1483497776091099,
|
|
"num_tokens": 7013815.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.512692880630493,
|
|
"epoch": 3.4207154297917777,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 6.0207,
|
|
"mean_token_accuracy": 0.1565142512321472,
|
|
"num_tokens": 7024212.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.381774616241455,
|
|
"epoch": 3.426054458088628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004268276179049292,
|
|
"loss": 6.0026,
|
|
"mean_token_accuracy": 0.15429308265447617,
|
|
"num_tokens": 7035167.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.468500185012817,
|
|
"epoch": 3.431393486385478,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004265153472320854,
|
|
"loss": 5.9676,
|
|
"mean_token_accuracy": 0.16137218475341797,
|
|
"num_tokens": 7045036.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.493806171417236,
|
|
"epoch": 3.436732514682328,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042620254164557137,
|
|
"loss": 6.0473,
|
|
"mean_token_accuracy": 0.1544056639075279,
|
|
"num_tokens": 7054938.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.475197839736938,
|
|
"epoch": 3.442071542979178,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000425889202249724,
|
|
"loss": 6.0545,
|
|
"mean_token_accuracy": 0.15177633464336396,
|
|
"num_tokens": 7065645.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.418945741653443,
|
|
"epoch": 3.447410571276028,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042557533015076456,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.1544858604669571,
|
|
"num_tokens": 7075909.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.486492872238159,
|
|
"epoch": 3.452749599572878,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042526092645679535,
|
|
"loss": 5.9918,
|
|
"mean_token_accuracy": 0.1586817979812622,
|
|
"num_tokens": 7085761.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.501900148391724,
|
|
"epoch": 3.458088627869728,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042494599227779523,
|
|
"loss": 6.053,
|
|
"mean_token_accuracy": 0.15420946329832078,
|
|
"num_tokens": 7096357.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.429144096374512,
|
|
"epoch": 3.4634276561665778,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042463052872561587,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.1528291255235672,
|
|
"num_tokens": 7107534.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.4680211544036865,
|
|
"epoch": 3.4687666844634277,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042431453691397787,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.14456742107868195,
|
|
"num_tokens": 7119186.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.5273964405059814,
|
|
"epoch": 3.4741057127602777,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004239980179584669,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.16706718355417252,
|
|
"num_tokens": 7130187.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.440040874481201,
|
|
"epoch": 3.4794447410571276,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004236809729765295,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.15689162760972977,
|
|
"num_tokens": 7140734.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.38389401435852,
|
|
"epoch": 3.4847837693539776,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042336340308746944,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.15397990345954896,
|
|
"num_tokens": 7151642.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.47389178276062,
|
|
"epoch": 3.4901227976508276,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004230453094124436,
|
|
"loss": 6.0409,
|
|
"mean_token_accuracy": 0.1492589846253395,
|
|
"num_tokens": 7162650.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.460881233215332,
|
|
"epoch": 3.4954618259476775,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042272669307445823,
|
|
"loss": 6.0319,
|
|
"mean_token_accuracy": 0.1465313658118248,
|
|
"num_tokens": 7174095.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.494899129867553,
|
|
"epoch": 3.5008008542445275,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042240755519836436,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.15974924415349961,
|
|
"num_tokens": 7185923.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.39954948425293,
|
|
"epoch": 3.5061398825413774,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042208789691085464,
|
|
"loss": 6.0127,
|
|
"mean_token_accuracy": 0.16205411553382873,
|
|
"num_tokens": 7197287.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.437206506729126,
|
|
"epoch": 3.5114789108382274,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000421767719340459,
|
|
"loss": 5.9969,
|
|
"mean_token_accuracy": 0.15230386182665825,
|
|
"num_tokens": 7208143.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.498978900909424,
|
|
"epoch": 3.5168179391350773,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004214470236175404,
|
|
"loss": 5.9591,
|
|
"mean_token_accuracy": 0.16153209805488586,
|
|
"num_tokens": 7218516.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.4734515190124515,
|
|
"epoch": 3.5221569674319273,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042112581087429133,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.16031200736761092,
|
|
"num_tokens": 7229066.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.440389013290405,
|
|
"epoch": 3.5274959957287773,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042080408224472946,
|
|
"loss": 6.02,
|
|
"mean_token_accuracy": 0.1543056882917881,
|
|
"num_tokens": 7239612.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.40929913520813,
|
|
"epoch": 3.532835024025627,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042048183886469386,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.15259570479393006,
|
|
"num_tokens": 7250266.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.427827405929565,
|
|
"epoch": 3.538174052322477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042015908187184077,
|
|
"loss": 6.0152,
|
|
"mean_token_accuracy": 0.14977363049983977,
|
|
"num_tokens": 7261183.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.471576929092407,
|
|
"epoch": 3.543513080619327,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041983581240563985,
|
|
"loss": 5.9753,
|
|
"mean_token_accuracy": 0.1603918567299843,
|
|
"num_tokens": 7271557.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.438574314117432,
|
|
"epoch": 3.5488521089161775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004195120316073698,
|
|
"loss": 6.0519,
|
|
"mean_token_accuracy": 0.15193243622779845,
|
|
"num_tokens": 7282540.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.431323099136352,
|
|
"epoch": 3.554191137213027,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041918774062011467,
|
|
"loss": 5.973,
|
|
"mean_token_accuracy": 0.15517610013484956,
|
|
"num_tokens": 7293568.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.432133626937866,
|
|
"epoch": 3.5595301655098774,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041886294058875975,
|
|
"loss": 6.0657,
|
|
"mean_token_accuracy": 0.14885200560092926,
|
|
"num_tokens": 7304745.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.448676490783692,
|
|
"epoch": 3.564869193806727,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004185376326599874,
|
|
"loss": 6.0468,
|
|
"mean_token_accuracy": 0.15006053894758226,
|
|
"num_tokens": 7316547.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.48239688873291,
|
|
"epoch": 3.5702082221035774,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041821181798227305,
|
|
"loss": 6.0789,
|
|
"mean_token_accuracy": 0.15182552486658096,
|
|
"num_tokens": 7328493.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.412248992919922,
|
|
"epoch": 3.575547250400427,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004178854977058812,
|
|
"loss": 6.0282,
|
|
"mean_token_accuracy": 0.15331946313381195,
|
|
"num_tokens": 7338831.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.476945543289185,
|
|
"epoch": 3.5808862786972773,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004175586729828614,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.14702386707067489,
|
|
"num_tokens": 7350884.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.464723682403564,
|
|
"epoch": 3.5862253069941272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004172313449670439,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.15223910957574843,
|
|
"num_tokens": 7360674.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.429730081558228,
|
|
"epoch": 3.591564335290977,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041690351481403606,
|
|
"loss": 6.0284,
|
|
"mean_token_accuracy": 0.1473098322749138,
|
|
"num_tokens": 7371965.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.518888187408447,
|
|
"epoch": 3.596903363587827,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004165751836812178,
|
|
"loss": 6.0972,
|
|
"mean_token_accuracy": 0.15113291889429092,
|
|
"num_tokens": 7383649.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.564088916778564,
|
|
"epoch": 3.602242391884677,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004162463527277378,
|
|
"loss": 6.0215,
|
|
"mean_token_accuracy": 0.1549656555056572,
|
|
"num_tokens": 7393692.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.46228060722351,
|
|
"epoch": 3.607581420181527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041591702311450924,
|
|
"loss": 6.0721,
|
|
"mean_token_accuracy": 0.14693051278591157,
|
|
"num_tokens": 7404149.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.439661455154419,
|
|
"epoch": 3.612920448478377,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004155871960042059,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.1573753163218498,
|
|
"num_tokens": 7415011.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.447901344299316,
|
|
"epoch": 3.618259476775227,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041525687256125786,
|
|
"loss": 6.0326,
|
|
"mean_token_accuracy": 0.1453842230141163,
|
|
"num_tokens": 7426297.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.3822489261627195,
|
|
"epoch": 3.623598505072077,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004149260539518475,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.15375188887119293,
|
|
"num_tokens": 7437718.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.435130977630616,
|
|
"epoch": 3.628937533368927,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041459474134390527,
|
|
"loss": 6.0035,
|
|
"mean_token_accuracy": 0.15174627155065537,
|
|
"num_tokens": 7448395.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.358313083648682,
|
|
"epoch": 3.634276561665777,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041426293590710587,
|
|
"loss": 5.9955,
|
|
"mean_token_accuracy": 0.15827126204967498,
|
|
"num_tokens": 7458374.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.455545091629029,
|
|
"epoch": 3.639615589962627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041393063881286346,
|
|
"loss": 6.0022,
|
|
"mean_token_accuracy": 0.15915190875530244,
|
|
"num_tokens": 7469386.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.489835023880005,
|
|
"epoch": 3.6449546182594768,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004135978512343284,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.1431760236620903,
|
|
"num_tokens": 7479788.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.448769283294678,
|
|
"epoch": 3.6502936465563267,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041326457434638246,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.1589450016617775,
|
|
"num_tokens": 7490329.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.481102228164673,
|
|
"epoch": 3.6556326748531767,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000412930809325635,
|
|
"loss": 6.0418,
|
|
"mean_token_accuracy": 0.14490215703845025,
|
|
"num_tokens": 7502206.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.423698377609253,
|
|
"epoch": 3.6609717031500266,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041259655735041847,
|
|
"loss": 6.054,
|
|
"mean_token_accuracy": 0.14937184154987335,
|
|
"num_tokens": 7513704.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.377822542190552,
|
|
"epoch": 3.6663107314468766,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041226181960078473,
|
|
"loss": 5.9491,
|
|
"mean_token_accuracy": 0.15650557577610016,
|
|
"num_tokens": 7524320.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.481767368316651,
|
|
"epoch": 3.6716497597437265,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004119265972585005,
|
|
"loss": 6.0168,
|
|
"mean_token_accuracy": 0.1564016103744507,
|
|
"num_tokens": 7533962.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.463397121429443,
|
|
"epoch": 3.6769887880405765,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041159089150704325,
|
|
"loss": 6.0754,
|
|
"mean_token_accuracy": 0.15199525505304337,
|
|
"num_tokens": 7545841.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.482867336273193,
|
|
"epoch": 3.6823278163374265,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041125470353159735,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.15769439339637756,
|
|
"num_tokens": 7555533.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.444973993301391,
|
|
"epoch": 3.6876668446342764,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041091803451904927,
|
|
"loss": 6.0022,
|
|
"mean_token_accuracy": 0.14563116431236267,
|
|
"num_tokens": 7566228.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.4181756496429445,
|
|
"epoch": 3.693005872931127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041058088565798403,
|
|
"loss": 6.0517,
|
|
"mean_token_accuracy": 0.15062558352947236,
|
|
"num_tokens": 7577694.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.527663326263427,
|
|
"epoch": 3.6983449012279763,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041024325813868065,
|
|
"loss": 6.0395,
|
|
"mean_token_accuracy": 0.1499044716358185,
|
|
"num_tokens": 7588141.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.487381887435913,
|
|
"epoch": 3.7036839295248267,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040990515315310794,
|
|
"loss": 6.0407,
|
|
"mean_token_accuracy": 0.1558864787220955,
|
|
"num_tokens": 7599831.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.454471158981323,
|
|
"epoch": 3.7090229578216762,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040956657189492046,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.15675635039806365,
|
|
"num_tokens": 7609567.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.458085393905639,
|
|
"epoch": 3.7143619861185266,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004092275155594541,
|
|
"loss": 6.0973,
|
|
"mean_token_accuracy": 0.1450599730014801,
|
|
"num_tokens": 7620722.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.480331087112427,
|
|
"epoch": 3.719701014415376,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004088879853437221,
|
|
"loss": 6.1177,
|
|
"mean_token_accuracy": 0.14813914149999619,
|
|
"num_tokens": 7633036.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.42266092300415,
|
|
"epoch": 3.7250400427122266,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040854798244641067,
|
|
"loss": 6.0263,
|
|
"mean_token_accuracy": 0.15134438276290893,
|
|
"num_tokens": 7642789.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.433456563949585,
|
|
"epoch": 3.7303790710090765,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004082075080678748,
|
|
"loss": 6.0446,
|
|
"mean_token_accuracy": 0.15298047214746474,
|
|
"num_tokens": 7654234.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.465288877487183,
|
|
"epoch": 3.7357180993059265,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040786656341013376,
|
|
"loss": 5.9805,
|
|
"mean_token_accuracy": 0.15464943945407866,
|
|
"num_tokens": 7665994.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.7357180993059265,
|
|
"eval_entropy": 6.249353146243405,
|
|
"eval_loss": 6.599543571472168,
|
|
"eval_mean_token_accuracy": 0.13020914061502978,
|
|
"eval_num_tokens": 7665994.0,
|
|
"eval_runtime": 2.0087,
|
|
"eval_samples_per_second": 1533.348,
|
|
"eval_steps_per_second": 191.669,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.412530136108399,
|
|
"epoch": 3.7410571276027764,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040752514967686747,
|
|
"loss": 5.9505,
|
|
"mean_token_accuracy": 0.15617771148681642,
|
|
"num_tokens": 7676813.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.359535884857178,
|
|
"epoch": 3.7463961558996264,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004071832680734118,
|
|
"loss": 5.9878,
|
|
"mean_token_accuracy": 0.1555604085326195,
|
|
"num_tokens": 7687775.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.460403823852539,
|
|
"epoch": 3.7517351841964763,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040684091980675426,
|
|
"loss": 6.0921,
|
|
"mean_token_accuracy": 0.14766517132520676,
|
|
"num_tokens": 7698909.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.422865295410157,
|
|
"epoch": 3.7570742124933263,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040649810608552994,
|
|
"loss": 5.9762,
|
|
"mean_token_accuracy": 0.15887688398361205,
|
|
"num_tokens": 7708783.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.464890670776367,
|
|
"epoch": 3.7624132407901762,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040615482812001726,
|
|
"loss": 6.0399,
|
|
"mean_token_accuracy": 0.15109720975160598,
|
|
"num_tokens": 7719010.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.455044603347778,
|
|
"epoch": 3.767752269087026,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004058110871221335,
|
|
"loss": 6.0723,
|
|
"mean_token_accuracy": 0.1478869929909706,
|
|
"num_tokens": 7730231.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.372013235092163,
|
|
"epoch": 3.773091297383876,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040546688430543074,
|
|
"loss": 5.9917,
|
|
"mean_token_accuracy": 0.15409810841083527,
|
|
"num_tokens": 7741656.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.463798379898071,
|
|
"epoch": 3.778430325680726,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040512222088509147,
|
|
"loss": 6.0021,
|
|
"mean_token_accuracy": 0.15271154344081878,
|
|
"num_tokens": 7751956.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.380717325210571,
|
|
"epoch": 3.783769353977576,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004047770980779242,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.16057718843221663,
|
|
"num_tokens": 7761944.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.344426584243775,
|
|
"epoch": 3.789108382274426,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004044315171023593,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.15577462315559387,
|
|
"num_tokens": 7771370.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.451844263076782,
|
|
"epoch": 3.794447410571276,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004040854791784448,
|
|
"loss": 6.1703,
|
|
"mean_token_accuracy": 0.1400241270661354,
|
|
"num_tokens": 7783615.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.545448112487793,
|
|
"epoch": 3.799786438868126,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004037389855278419,
|
|
"loss": 6.0275,
|
|
"mean_token_accuracy": 0.15665434598922728,
|
|
"num_tokens": 7794236.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.422525453567505,
|
|
"epoch": 3.805125467164976,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040339203737382065,
|
|
"loss": 6.0238,
|
|
"mean_token_accuracy": 0.15270079523324967,
|
|
"num_tokens": 7806158.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.394365024566651,
|
|
"epoch": 3.810464495461826,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004030446359412556,
|
|
"loss": 5.9801,
|
|
"mean_token_accuracy": 0.15192456245422364,
|
|
"num_tokens": 7818283.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.483584976196289,
|
|
"epoch": 3.815803523758676,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004026967824566218,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.14997316673398017,
|
|
"num_tokens": 7828683.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.420600032806396,
|
|
"epoch": 3.8211425520555258,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004023484781479902,
|
|
"loss": 5.9998,
|
|
"mean_token_accuracy": 0.15755953192710875,
|
|
"num_tokens": 7839116.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.454991292953491,
|
|
"epoch": 3.8264815803523757,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 6.0556,
|
|
"mean_token_accuracy": 0.15305356681346893,
|
|
"num_tokens": 7850300.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.454456949234009,
|
|
"epoch": 3.8318206086492257,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004016505219789703,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.15290531367063523,
|
|
"num_tokens": 7861432.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.36416540145874,
|
|
"epoch": 3.837159636946076,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004013008725826644,
|
|
"loss": 6.0253,
|
|
"mean_token_accuracy": 0.15670392215251921,
|
|
"num_tokens": 7872472.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.47998948097229,
|
|
"epoch": 3.8424986652429256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004009507772905167,
|
|
"loss": 6.0442,
|
|
"mean_token_accuracy": 0.15383006036281585,
|
|
"num_tokens": 7883554.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.492900037765503,
|
|
"epoch": 3.847837693539776,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040060023733851254,
|
|
"loss": 6.0344,
|
|
"mean_token_accuracy": 0.1482291743159294,
|
|
"num_tokens": 7894853.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.387452697753906,
|
|
"epoch": 3.8531767218366255,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040024925396420707,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.14481948912143708,
|
|
"num_tokens": 7905459.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.390704679489136,
|
|
"epoch": 3.858515750133476,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039989782840672123,
|
|
"loss": 5.9617,
|
|
"mean_token_accuracy": 0.15468235164880753,
|
|
"num_tokens": 7916646.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.4032450199127195,
|
|
"epoch": 3.8638547784303254,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995459619067366,
|
|
"loss": 6.0086,
|
|
"mean_token_accuracy": 0.15670197159051896,
|
|
"num_tokens": 7928119.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.462900686264038,
|
|
"epoch": 3.869193806727176,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039919365570649187,
|
|
"loss": 6.1133,
|
|
"mean_token_accuracy": 0.1466163083910942,
|
|
"num_tokens": 7939295.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.471037149429321,
|
|
"epoch": 3.874532835024026,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039884091104977774,
|
|
"loss": 5.9457,
|
|
"mean_token_accuracy": 0.15363116711378097,
|
|
"num_tokens": 7950094.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.446648645401001,
|
|
"epoch": 3.8798718633208757,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039848772918193315,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.16060514748096466,
|
|
"num_tokens": 7960767.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.353861904144287,
|
|
"epoch": 3.8852108916177257,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003981341113498404,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.1542203426361084,
|
|
"num_tokens": 7971660.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.439627313613892,
|
|
"epoch": 3.8905499199145757,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003977800588019209,
|
|
"loss": 6.0221,
|
|
"mean_token_accuracy": 0.15386180579662323,
|
|
"num_tokens": 7983337.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.4260170459747314,
|
|
"epoch": 3.8958889482114256,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003974255727881307,
|
|
"loss": 6.0604,
|
|
"mean_token_accuracy": 0.15081475675106049,
|
|
"num_tokens": 7994540.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.435766696929932,
|
|
"epoch": 3.9012279765082756,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003970706545599566,
|
|
"loss": 5.965,
|
|
"mean_token_accuracy": 0.15007809400558472,
|
|
"num_tokens": 8005396.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.3866301536560055,
|
|
"epoch": 3.9065670048051255,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039671530537041094,
|
|
"loss": 5.9777,
|
|
"mean_token_accuracy": 0.15678471177816392,
|
|
"num_tokens": 8015411.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.396858072280883,
|
|
"epoch": 3.9119060331019755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039635952647402756,
|
|
"loss": 5.9542,
|
|
"mean_token_accuracy": 0.1602507583796978,
|
|
"num_tokens": 8025720.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.39530029296875,
|
|
"epoch": 3.9172450613988254,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003960033191268574,
|
|
"loss": 6.0026,
|
|
"mean_token_accuracy": 0.15195597410202027,
|
|
"num_tokens": 8036081.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.354433155059814,
|
|
"epoch": 3.9225840896956754,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003956466845864641,
|
|
"loss": 5.8865,
|
|
"mean_token_accuracy": 0.15650326758623123,
|
|
"num_tokens": 8046714.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.3409348487854,
|
|
"epoch": 3.9279231179925254,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003952896241119194,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.1585497096180916,
|
|
"num_tokens": 8057174.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.35273060798645,
|
|
"epoch": 3.9332621462893753,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003949321389637986,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.15953728854656218,
|
|
"num_tokens": 8067697.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.431153202056885,
|
|
"epoch": 3.9386011745862253,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003945742304041766,
|
|
"loss": 5.9975,
|
|
"mean_token_accuracy": 0.15753055959939957,
|
|
"num_tokens": 8077800.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.457331705093384,
|
|
"epoch": 3.9439402028830752,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003942158996966229,
|
|
"loss": 6.0855,
|
|
"mean_token_accuracy": 0.14850110039114953,
|
|
"num_tokens": 8090319.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.439395999908447,
|
|
"epoch": 3.949279231179925,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003938571481061975,
|
|
"loss": 6.0262,
|
|
"mean_token_accuracy": 0.15176256448030473,
|
|
"num_tokens": 8100498.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.383943176269531,
|
|
"epoch": 3.954618259476775,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039349797689944613,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.15272487998008727,
|
|
"num_tokens": 8110506.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.435845994949341,
|
|
"epoch": 3.959957287773625,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039313838734439605,
|
|
"loss": 5.9817,
|
|
"mean_token_accuracy": 0.15748539045453072,
|
|
"num_tokens": 8121999.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.461563158035278,
|
|
"epoch": 3.965296316070475,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039277838071055145,
|
|
"loss": 6.044,
|
|
"mean_token_accuracy": 0.15032595545053482,
|
|
"num_tokens": 8132884.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.446934556961059,
|
|
"epoch": 3.970635344367325,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000392417958268889,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.1556297719478607,
|
|
"num_tokens": 8143344.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.390069246292114,
|
|
"epoch": 3.975974372664175,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039205712129185337,
|
|
"loss": 5.9765,
|
|
"mean_token_accuracy": 0.15745232701301576,
|
|
"num_tokens": 8154658.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.385173702239991,
|
|
"epoch": 3.981313400961025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003916958710533528,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.156840480864048,
|
|
"num_tokens": 8166057.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.434129476547241,
|
|
"epoch": 3.986652429257875,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003913342088287541,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.15821003764867783,
|
|
"num_tokens": 8176470.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.455521726608277,
|
|
"epoch": 3.9919914575547253,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000390972135894879,
|
|
"loss": 6.0344,
|
|
"mean_token_accuracy": 0.15137473344802857,
|
|
"num_tokens": 8186458.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.492480516433716,
|
|
"epoch": 3.997330485851575,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039060965352999926,
|
|
"loss": 6.1193,
|
|
"mean_token_accuracy": 0.14308308064937592,
|
|
"num_tokens": 8197826.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.441369639502631,
|
|
"epoch": 4.00213561131874,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003902467630138317,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.15558469461070168,
|
|
"num_tokens": 8208809.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.367521619796753,
|
|
"epoch": 4.00747463961559,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038988346562753435,
|
|
"loss": 5.674,
|
|
"mean_token_accuracy": 0.1742790684103966,
|
|
"num_tokens": 8218740.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.356276035308838,
|
|
"epoch": 4.0128136679124395,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038951976265370176,
|
|
"loss": 5.6682,
|
|
"mean_token_accuracy": 0.1751902312040329,
|
|
"num_tokens": 8230360.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.419689798355103,
|
|
"epoch": 4.01815269620929,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038915565537636003,
|
|
"loss": 5.6982,
|
|
"mean_token_accuracy": 0.1685792848467827,
|
|
"num_tokens": 8240886.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.367136859893799,
|
|
"epoch": 4.0234917245061395,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000388791145080963,
|
|
"loss": 5.648,
|
|
"mean_token_accuracy": 0.17499402165412903,
|
|
"num_tokens": 8252103.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.4103758335113525,
|
|
"epoch": 4.02883075280299,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003884262330543872,
|
|
"loss": 5.7569,
|
|
"mean_token_accuracy": 0.16168682277202606,
|
|
"num_tokens": 8262890.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.357025671005249,
|
|
"epoch": 4.03416978109984,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038806092058492727,
|
|
"loss": 5.6842,
|
|
"mean_token_accuracy": 0.17419513165950776,
|
|
"num_tokens": 8273562.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.451374292373657,
|
|
"epoch": 4.03950880939669,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003876952089622918,
|
|
"loss": 5.7392,
|
|
"mean_token_accuracy": 0.159723262488842,
|
|
"num_tokens": 8285104.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.4013794422149655,
|
|
"epoch": 4.04484783769354,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003873290994775984,
|
|
"loss": 5.736,
|
|
"mean_token_accuracy": 0.16341031938791276,
|
|
"num_tokens": 8296760.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.383622455596924,
|
|
"epoch": 4.05018686599039,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038696259342336966,
|
|
"loss": 5.7409,
|
|
"mean_token_accuracy": 0.1678796350955963,
|
|
"num_tokens": 8306929.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.384306144714356,
|
|
"epoch": 4.05552589428724,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038659569209352766,
|
|
"loss": 5.7687,
|
|
"mean_token_accuracy": 0.1669505715370178,
|
|
"num_tokens": 8316736.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.400050210952759,
|
|
"epoch": 4.06086492258409,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038622839678339034,
|
|
"loss": 5.7442,
|
|
"mean_token_accuracy": 0.17051567882299423,
|
|
"num_tokens": 8327688.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.379993724822998,
|
|
"epoch": 4.06620395088094,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00038586070878966663,
|
|
"loss": 5.6541,
|
|
"mean_token_accuracy": 0.17766587436199188,
|
|
"num_tokens": 8337986.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.322160434722901,
|
|
"epoch": 4.0715429791777895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003854926294104514,
|
|
"loss": 5.664,
|
|
"mean_token_accuracy": 0.17622167468070984,
|
|
"num_tokens": 8348473.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.410195541381836,
|
|
"epoch": 4.07688200747464,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003851241599452218,
|
|
"loss": 5.6823,
|
|
"mean_token_accuracy": 0.16867805123329163,
|
|
"num_tokens": 8359563.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.374693393707275,
|
|
"epoch": 4.082221035771489,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038475530169483185,
|
|
"loss": 5.7864,
|
|
"mean_token_accuracy": 0.16349997520446777,
|
|
"num_tokens": 8370965.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.359370851516724,
|
|
"epoch": 4.08756006406834,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038438605596150807,
|
|
"loss": 5.7789,
|
|
"mean_token_accuracy": 0.16721495538949965,
|
|
"num_tokens": 8382606.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.406767559051514,
|
|
"epoch": 4.092899092365189,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038401642404884513,
|
|
"loss": 5.7299,
|
|
"mean_token_accuracy": 0.17507466077804565,
|
|
"num_tokens": 8392618.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.3540387630462645,
|
|
"epoch": 4.09823812066204,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038364640726180096,
|
|
"loss": 5.6719,
|
|
"mean_token_accuracy": 0.16978386491537095,
|
|
"num_tokens": 8403210.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.389392566680908,
|
|
"epoch": 4.103577148958889,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038327600690669244,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.1627689242362976,
|
|
"num_tokens": 8413838.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.315148067474365,
|
|
"epoch": 4.10891617725574,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038290522429119034,
|
|
"loss": 5.6946,
|
|
"mean_token_accuracy": 0.17259662300348283,
|
|
"num_tokens": 8425282.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.347264814376831,
|
|
"epoch": 4.114255205552589,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038253406072431525,
|
|
"loss": 5.6852,
|
|
"mean_token_accuracy": 0.17276675552129744,
|
|
"num_tokens": 8436762.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.399221420288086,
|
|
"epoch": 4.11959423384944,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038216251751643245,
|
|
"loss": 5.7487,
|
|
"mean_token_accuracy": 0.17090824246406555,
|
|
"num_tokens": 8448636.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.338019895553589,
|
|
"epoch": 4.124933262146289,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003817905959792477,
|
|
"loss": 5.6623,
|
|
"mean_token_accuracy": 0.17457616031169892,
|
|
"num_tokens": 8459730.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.338306093215943,
|
|
"epoch": 4.1302722904431395,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003814182974258023,
|
|
"loss": 5.6725,
|
|
"mean_token_accuracy": 0.17892919182777406,
|
|
"num_tokens": 8470628.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.307374906539917,
|
|
"epoch": 4.135611318739989,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038104562317046873,
|
|
"loss": 5.6806,
|
|
"mean_token_accuracy": 0.17068182975053786,
|
|
"num_tokens": 8480785.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.298294639587402,
|
|
"epoch": 4.140950347036839,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003806725745289455,
|
|
"loss": 5.634,
|
|
"mean_token_accuracy": 0.1782844215631485,
|
|
"num_tokens": 8491596.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.351423168182373,
|
|
"epoch": 4.146289375333689,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038029915281825334,
|
|
"loss": 5.7261,
|
|
"mean_token_accuracy": 0.17444997280836105,
|
|
"num_tokens": 8502458.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.383857011795044,
|
|
"epoch": 4.151628403630539,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037992535935672974,
|
|
"loss": 5.69,
|
|
"mean_token_accuracy": 0.17884884476661683,
|
|
"num_tokens": 8513018.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.426295518875122,
|
|
"epoch": 4.156967431927389,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037955119546402474,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.17245568037033082,
|
|
"num_tokens": 8523757.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.244861793518067,
|
|
"epoch": 4.162306460224239,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003791766624610961,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.17205275297164918,
|
|
"num_tokens": 8535446.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.25617995262146,
|
|
"epoch": 4.167645488521089,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003788017616702048,
|
|
"loss": 5.6233,
|
|
"mean_token_accuracy": 0.18538019359111785,
|
|
"num_tokens": 8545868.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.315582132339477,
|
|
"epoch": 4.172984516817939,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003784264944149102,
|
|
"loss": 5.6834,
|
|
"mean_token_accuracy": 0.17073606699705124,
|
|
"num_tokens": 8556465.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.362593936920166,
|
|
"epoch": 4.1783235451147895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003780508620200655,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.17016505002975463,
|
|
"num_tokens": 8568227.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.335514879226684,
|
|
"epoch": 4.183662573411639,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003776748658118128,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.17090839445590972,
|
|
"num_tokens": 8579309.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.341404008865356,
|
|
"epoch": 4.1890016017084895,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037729850711757903,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.17466934472322465,
|
|
"num_tokens": 8590487.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.418704223632813,
|
|
"epoch": 4.194340630005339,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003769217872660703,
|
|
"loss": 5.7346,
|
|
"mean_token_accuracy": 0.17014706283807754,
|
|
"num_tokens": 8600940.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.312211322784424,
|
|
"epoch": 4.199679658302189,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037654470758726816,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.1879582181572914,
|
|
"num_tokens": 8611322.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.290743589401245,
|
|
"epoch": 4.205018686599039,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037616726941242445,
|
|
"loss": 5.7308,
|
|
"mean_token_accuracy": 0.168832990527153,
|
|
"num_tokens": 8622055.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.33185224533081,
|
|
"epoch": 4.210357714895889,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003757894740740565,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.1679001972079277,
|
|
"num_tokens": 8633515.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.292371559143066,
|
|
"epoch": 4.215696743192739,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003754113229059426,
|
|
"loss": 5.6713,
|
|
"mean_token_accuracy": 0.17913059443235396,
|
|
"num_tokens": 8644199.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.2938971519470215,
|
|
"epoch": 4.221035771489589,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003750328172431176,
|
|
"loss": 5.7341,
|
|
"mean_token_accuracy": 0.16677308976650237,
|
|
"num_tokens": 8655025.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.300493478775024,
|
|
"epoch": 4.226374799786439,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003746539584218673,
|
|
"loss": 5.7044,
|
|
"mean_token_accuracy": 0.1770304784178734,
|
|
"num_tokens": 8666251.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.288543462753296,
|
|
"epoch": 4.231713828083289,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.17697502821683883,
|
|
"num_tokens": 8676727.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.354572534561157,
|
|
"epoch": 4.237052856380139,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000373895186655465,
|
|
"loss": 5.7337,
|
|
"mean_token_accuracy": 0.17722104340791703,
|
|
"num_tokens": 8687498.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.396370840072632,
|
|
"epoch": 4.242391884676989,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037351527638910035,
|
|
"loss": 5.6905,
|
|
"mean_token_accuracy": 0.1682073011994362,
|
|
"num_tokens": 8698505.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.378250169754028,
|
|
"epoch": 4.2477309129738385,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003731350183218755,
|
|
"loss": 5.7976,
|
|
"mean_token_accuracy": 0.16625124961137772,
|
|
"num_tokens": 8709349.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.36450982093811,
|
|
"epoch": 4.253069941270689,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003727544137962637,
|
|
"loss": 5.6502,
|
|
"mean_token_accuracy": 0.16934423595666886,
|
|
"num_tokens": 8719515.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.346686744689942,
|
|
"epoch": 4.2584089695675384,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003723734641559607,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.16725968718528747,
|
|
"num_tokens": 8731030.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.3356071472167965,
|
|
"epoch": 4.263747997864389,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003719921707458809,
|
|
"loss": 5.7235,
|
|
"mean_token_accuracy": 0.16923437565565108,
|
|
"num_tokens": 8742776.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.330284976959229,
|
|
"epoch": 4.269087026161238,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003716105349121523,
|
|
"loss": 5.6449,
|
|
"mean_token_accuracy": 0.17387565523386,
|
|
"num_tokens": 8753994.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.269087026161238,
|
|
"eval_entropy": 6.115814966969676,
|
|
"eval_loss": 6.588654041290283,
|
|
"eval_mean_token_accuracy": 0.13136930310880984,
|
|
"eval_num_tokens": 8753994.0,
|
|
"eval_runtime": 1.8176,
|
|
"eval_samples_per_second": 1694.53,
|
|
"eval_steps_per_second": 211.816,
|
|
"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
|
|
}
|