6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.585308750365818,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742545986175537,
|
|
"epoch": 0.0009755145839430299,
|
|
"grad_norm": 4.65625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8379,
|
|
"mean_token_accuracy": 0.00014434070908464491,
|
|
"num_tokens": 17120.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742573165893555,
|
|
"epoch": 0.0019510291678860598,
|
|
"grad_norm": 4.96875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.8311,
|
|
"mean_token_accuracy": 0.00011998222325928509,
|
|
"num_tokens": 34182.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742534160614014,
|
|
"epoch": 0.0029265437518290896,
|
|
"grad_norm": 5.0,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.8027,
|
|
"mean_token_accuracy": 0.00012189020635560155,
|
|
"num_tokens": 51205.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742509460449218,
|
|
"epoch": 0.0039020583357721197,
|
|
"grad_norm": 5.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.7494,
|
|
"mean_token_accuracy": 0.0005705497809685766,
|
|
"num_tokens": 67722.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742345905303955,
|
|
"epoch": 0.00487757291971515,
|
|
"grad_norm": 4.4375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6494,
|
|
"mean_token_accuracy": 0.02957919198088348,
|
|
"num_tokens": 84230.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741327095031739,
|
|
"epoch": 0.005853087503658179,
|
|
"grad_norm": 3.75,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5266,
|
|
"mean_token_accuracy": 0.061843523755669594,
|
|
"num_tokens": 100967.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.7347092628479,
|
|
"epoch": 0.00682860208760121,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3637,
|
|
"mean_token_accuracy": 0.06739617437124253,
|
|
"num_tokens": 117450.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.70640697479248,
|
|
"epoch": 0.007804116671544239,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2606,
|
|
"mean_token_accuracy": 0.06366759836673737,
|
|
"num_tokens": 135328.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.657337093353272,
|
|
"epoch": 0.00877963125548727,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1636,
|
|
"mean_token_accuracy": 0.06608470231294632,
|
|
"num_tokens": 152571.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.620244789123536,
|
|
"epoch": 0.0097551458394303,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0949,
|
|
"mean_token_accuracy": 0.0651657074689865,
|
|
"num_tokens": 170216.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.61258020401001,
|
|
"epoch": 0.010730660423373329,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.015,
|
|
"mean_token_accuracy": 0.07096394337713718,
|
|
"num_tokens": 186513.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.613809299468993,
|
|
"epoch": 0.011706175007316359,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.0117,
|
|
"mean_token_accuracy": 0.06269222758710384,
|
|
"num_tokens": 202600.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.618330478668213,
|
|
"epoch": 0.01268168959125939,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.9438,
|
|
"mean_token_accuracy": 0.07208959870040417,
|
|
"num_tokens": 220270.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.610760307312011,
|
|
"epoch": 0.01365720417520242,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8933,
|
|
"mean_token_accuracy": 0.06949375495314598,
|
|
"num_tokens": 236810.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.593310832977295,
|
|
"epoch": 0.014632718759145449,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.8518,
|
|
"mean_token_accuracy": 0.07177168019115925,
|
|
"num_tokens": 252559.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.562477207183838,
|
|
"epoch": 0.015608233343088479,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7687,
|
|
"mean_token_accuracy": 0.07241420224308967,
|
|
"num_tokens": 268507.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.535469722747802,
|
|
"epoch": 0.01658374792703151,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.7325,
|
|
"mean_token_accuracy": 0.0762402530759573,
|
|
"num_tokens": 284776.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.517255306243896,
|
|
"epoch": 0.01755926251097454,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.7206,
|
|
"mean_token_accuracy": 0.07280966117978097,
|
|
"num_tokens": 300937.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.505683422088623,
|
|
"epoch": 0.01853477709491757,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.654,
|
|
"mean_token_accuracy": 0.07415377609431743,
|
|
"num_tokens": 317464.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.483844089508057,
|
|
"epoch": 0.0195102916788606,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.5346,
|
|
"mean_token_accuracy": 0.0789559468626976,
|
|
"num_tokens": 334022.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.463724327087402,
|
|
"epoch": 0.02048580626280363,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.4998,
|
|
"mean_token_accuracy": 0.07889014333486558,
|
|
"num_tokens": 350545.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.37066593170166,
|
|
"epoch": 0.021461320846746658,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.4199,
|
|
"mean_token_accuracy": 0.08481870368123054,
|
|
"num_tokens": 368878.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.354177665710449,
|
|
"epoch": 0.022436835430689687,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.3886,
|
|
"mean_token_accuracy": 0.0817614309489727,
|
|
"num_tokens": 388381.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.36144495010376,
|
|
"epoch": 0.023412350014632717,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3266,
|
|
"mean_token_accuracy": 0.07798456773161888,
|
|
"num_tokens": 403727.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.321478271484375,
|
|
"epoch": 0.02438786459857575,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2543,
|
|
"mean_token_accuracy": 0.07955141216516495,
|
|
"num_tokens": 419046.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.201090621948243,
|
|
"epoch": 0.02536337918251878,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.1187,
|
|
"mean_token_accuracy": 0.08797711431980133,
|
|
"num_tokens": 435186.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.184612846374511,
|
|
"epoch": 0.02633889376646181,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.0877,
|
|
"mean_token_accuracy": 0.08292194977402687,
|
|
"num_tokens": 451552.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.037136459350586,
|
|
"epoch": 0.02731440835040484,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.0324,
|
|
"mean_token_accuracy": 0.08637636378407479,
|
|
"num_tokens": 468006.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.057977771759033,
|
|
"epoch": 0.02828992293434787,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.9508,
|
|
"mean_token_accuracy": 0.08355718627572059,
|
|
"num_tokens": 484381.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.876834678649903,
|
|
"epoch": 0.029265437518290898,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.93,
|
|
"mean_token_accuracy": 0.08421113938093186,
|
|
"num_tokens": 501309.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.873361778259277,
|
|
"epoch": 0.030240952102233928,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.8471,
|
|
"mean_token_accuracy": 0.07950809225440025,
|
|
"num_tokens": 518290.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.737128734588623,
|
|
"epoch": 0.031216466686176957,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.7745,
|
|
"mean_token_accuracy": 0.08451616615056992,
|
|
"num_tokens": 533700.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.578966617584229,
|
|
"epoch": 0.03219198127011999,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.6201,
|
|
"mean_token_accuracy": 0.08652802109718323,
|
|
"num_tokens": 548935.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.524422359466552,
|
|
"epoch": 0.03316749585406302,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.6444,
|
|
"mean_token_accuracy": 0.08372659981250763,
|
|
"num_tokens": 565869.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.335462665557861,
|
|
"epoch": 0.03414301043800605,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.6507,
|
|
"mean_token_accuracy": 0.08460443243384361,
|
|
"num_tokens": 583065.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.36280517578125,
|
|
"epoch": 0.03511852502194908,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.6329,
|
|
"mean_token_accuracy": 0.08482759669423104,
|
|
"num_tokens": 599875.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.173808383941651,
|
|
"epoch": 0.03609403960589211,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.5743,
|
|
"mean_token_accuracy": 0.08783280774950981,
|
|
"num_tokens": 616296.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.197137355804443,
|
|
"epoch": 0.03706955418983514,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.5285,
|
|
"mean_token_accuracy": 0.08100056424736976,
|
|
"num_tokens": 632548.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.10781021118164,
|
|
"epoch": 0.03804506877377817,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.5301,
|
|
"mean_token_accuracy": 0.08742500841617584,
|
|
"num_tokens": 650395.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 9.043368911743164,
|
|
"epoch": 0.0390205833577212,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.5218,
|
|
"mean_token_accuracy": 0.09499027654528618,
|
|
"num_tokens": 668453.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 9.030275917053222,
|
|
"epoch": 0.03999609794166423,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.4423,
|
|
"mean_token_accuracy": 0.08772219642996788,
|
|
"num_tokens": 683914.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.929483890533447,
|
|
"epoch": 0.04097161252560726,
|
|
"grad_norm": 0.66796875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.5108,
|
|
"mean_token_accuracy": 0.08664508312940597,
|
|
"num_tokens": 700480.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.955109214782714,
|
|
"epoch": 0.041947127109550286,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4882,
|
|
"mean_token_accuracy": 0.08451240882277489,
|
|
"num_tokens": 716604.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.781434345245362,
|
|
"epoch": 0.042922641693493316,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.4715,
|
|
"mean_token_accuracy": 0.08864145204424859,
|
|
"num_tokens": 732697.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.88281774520874,
|
|
"epoch": 0.043898156277436345,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.5344,
|
|
"mean_token_accuracy": 0.08214838653802872,
|
|
"num_tokens": 749072.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.81209487915039,
|
|
"epoch": 0.044873670861379375,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.4873,
|
|
"mean_token_accuracy": 0.08687537536025047,
|
|
"num_tokens": 767826.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.874574089050293,
|
|
"epoch": 0.045849185445322405,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.4097,
|
|
"mean_token_accuracy": 0.09442409723997117,
|
|
"num_tokens": 786863.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.778143787384034,
|
|
"epoch": 0.046824700029265434,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3929,
|
|
"mean_token_accuracy": 0.10005151331424714,
|
|
"num_tokens": 803925.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.731143760681153,
|
|
"epoch": 0.04780021461320847,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.4637,
|
|
"mean_token_accuracy": 0.09069854766130447,
|
|
"num_tokens": 822088.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.784804248809815,
|
|
"epoch": 0.0487757291971515,
|
|
"grad_norm": 0.671875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.4703,
|
|
"mean_token_accuracy": 0.08817133978009224,
|
|
"num_tokens": 837991.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.71630163192749,
|
|
"epoch": 0.04975124378109453,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3759,
|
|
"mean_token_accuracy": 0.09599229618906975,
|
|
"num_tokens": 855593.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.778234958648682,
|
|
"epoch": 0.05072675836503756,
|
|
"grad_norm": 0.6953125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3926,
|
|
"mean_token_accuracy": 0.08905554860830307,
|
|
"num_tokens": 873696.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.698340606689452,
|
|
"epoch": 0.05170227294898059,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3648,
|
|
"mean_token_accuracy": 0.09724866859614849,
|
|
"num_tokens": 890023.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.724142837524415,
|
|
"epoch": 0.05267778753292362,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.499,
|
|
"mean_token_accuracy": 0.08494378551840782,
|
|
"num_tokens": 906492.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.658268070220947,
|
|
"epoch": 0.05365330211686665,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.4326,
|
|
"mean_token_accuracy": 0.0881828211247921,
|
|
"num_tokens": 921837.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.749652004241943,
|
|
"epoch": 0.05462881670080968,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.3803,
|
|
"mean_token_accuracy": 0.10005034953355789,
|
|
"num_tokens": 938771.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.66503381729126,
|
|
"epoch": 0.05560433128475271,
|
|
"grad_norm": 0.65234375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.388,
|
|
"mean_token_accuracy": 0.08957169800996781,
|
|
"num_tokens": 954562.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.760249996185303,
|
|
"epoch": 0.05657984586869574,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.3309,
|
|
"mean_token_accuracy": 0.09739773571491242,
|
|
"num_tokens": 973700.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.610127925872803,
|
|
"epoch": 0.05755536045263877,
|
|
"grad_norm": 0.6328125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.3841,
|
|
"mean_token_accuracy": 0.09089891463518143,
|
|
"num_tokens": 990688.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.602487468719483,
|
|
"epoch": 0.058530875036581796,
|
|
"grad_norm": 0.65234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.4543,
|
|
"mean_token_accuracy": 0.08643004968762398,
|
|
"num_tokens": 1008144.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.684108543395997,
|
|
"epoch": 0.059506389620524826,
|
|
"grad_norm": 0.62109375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.4075,
|
|
"mean_token_accuracy": 0.08840584978461266,
|
|
"num_tokens": 1027379.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.618590164184571,
|
|
"epoch": 0.060481904204467855,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3946,
|
|
"mean_token_accuracy": 0.09471525549888611,
|
|
"num_tokens": 1043999.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.71361255645752,
|
|
"epoch": 0.061457418788410885,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.3943,
|
|
"mean_token_accuracy": 0.08788567930459976,
|
|
"num_tokens": 1062237.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.668982696533202,
|
|
"epoch": 0.062432933372353915,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.3439,
|
|
"mean_token_accuracy": 0.09140986129641533,
|
|
"num_tokens": 1078092.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.612729835510255,
|
|
"epoch": 0.06340844795629695,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.3807,
|
|
"mean_token_accuracy": 0.09364238902926444,
|
|
"num_tokens": 1095124.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.613638973236084,
|
|
"epoch": 0.06438396254023998,
|
|
"grad_norm": 0.65625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.3699,
|
|
"mean_token_accuracy": 0.08994951322674752,
|
|
"num_tokens": 1112775.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.583961772918702,
|
|
"epoch": 0.06535947712418301,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.306,
|
|
"mean_token_accuracy": 0.09773509949445724,
|
|
"num_tokens": 1129726.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.63847484588623,
|
|
"epoch": 0.06633499170812604,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.3582,
|
|
"mean_token_accuracy": 0.09658835083246231,
|
|
"num_tokens": 1144034.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.646141052246094,
|
|
"epoch": 0.06731050629206907,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.3001,
|
|
"mean_token_accuracy": 0.09566695243120193,
|
|
"num_tokens": 1159633.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.552906799316407,
|
|
"epoch": 0.0682860208760121,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2566,
|
|
"mean_token_accuracy": 0.09137818142771721,
|
|
"num_tokens": 1175350.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.47085542678833,
|
|
"epoch": 0.06926153545995513,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.2038,
|
|
"mean_token_accuracy": 0.09639394730329513,
|
|
"num_tokens": 1193121.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.64188289642334,
|
|
"epoch": 0.07023705004389816,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.352,
|
|
"mean_token_accuracy": 0.09170110374689103,
|
|
"num_tokens": 1209896.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.56321315765381,
|
|
"epoch": 0.07121256462784119,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.3321,
|
|
"mean_token_accuracy": 0.09032948762178421,
|
|
"num_tokens": 1227011.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.49931468963623,
|
|
"epoch": 0.07218807921178422,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.2586,
|
|
"mean_token_accuracy": 0.0936131976544857,
|
|
"num_tokens": 1241595.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.551326656341553,
|
|
"epoch": 0.07316359379572725,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.2809,
|
|
"mean_token_accuracy": 0.08881813138723374,
|
|
"num_tokens": 1256919.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.576890087127685,
|
|
"epoch": 0.07413910837967028,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.2696,
|
|
"mean_token_accuracy": 0.10841422453522682,
|
|
"num_tokens": 1274344.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.525260543823242,
|
|
"epoch": 0.0751146229636133,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.2441,
|
|
"mean_token_accuracy": 0.09685440510511398,
|
|
"num_tokens": 1290252.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.542247867584228,
|
|
"epoch": 0.07609013754755634,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.3092,
|
|
"mean_token_accuracy": 0.09308558702468872,
|
|
"num_tokens": 1306987.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.586502170562744,
|
|
"epoch": 0.07706565213149937,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.3282,
|
|
"mean_token_accuracy": 0.0875322088599205,
|
|
"num_tokens": 1323031.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.52306661605835,
|
|
"epoch": 0.0780411667154424,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.2677,
|
|
"mean_token_accuracy": 0.09020323976874352,
|
|
"num_tokens": 1339592.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.584537792205811,
|
|
"epoch": 0.07901668129938542,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.2469,
|
|
"mean_token_accuracy": 0.0998822957277298,
|
|
"num_tokens": 1358131.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.492807197570801,
|
|
"epoch": 0.07999219588332845,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.1846,
|
|
"mean_token_accuracy": 0.10047675669193268,
|
|
"num_tokens": 1375705.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.449185466766357,
|
|
"epoch": 0.08096771046727148,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.2014,
|
|
"mean_token_accuracy": 0.09942332729697227,
|
|
"num_tokens": 1392736.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.513163948059082,
|
|
"epoch": 0.08194322505121451,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.2322,
|
|
"mean_token_accuracy": 0.08992344737052918,
|
|
"num_tokens": 1408536.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.381054210662843,
|
|
"epoch": 0.08291873963515754,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.2024,
|
|
"mean_token_accuracy": 0.09349315315485,
|
|
"num_tokens": 1425423.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.559225177764892,
|
|
"epoch": 0.08389425421910057,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.2133,
|
|
"mean_token_accuracy": 0.09476829543709755,
|
|
"num_tokens": 1442420.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.574479293823241,
|
|
"epoch": 0.0848697688030436,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.1947,
|
|
"mean_token_accuracy": 0.09402725324034691,
|
|
"num_tokens": 1458835.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.469829273223876,
|
|
"epoch": 0.08584528338698663,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.2763,
|
|
"mean_token_accuracy": 0.09046141654253007,
|
|
"num_tokens": 1475563.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.52817611694336,
|
|
"epoch": 0.08682079797092966,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.2133,
|
|
"mean_token_accuracy": 0.10233243256807327,
|
|
"num_tokens": 1492687.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.53304328918457,
|
|
"epoch": 0.08779631255487269,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.2363,
|
|
"mean_token_accuracy": 0.09168101623654365,
|
|
"num_tokens": 1508394.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.450295543670654,
|
|
"epoch": 0.08877182713881572,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.2079,
|
|
"mean_token_accuracy": 0.10227799862623214,
|
|
"num_tokens": 1527112.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.415885162353515,
|
|
"epoch": 0.08974734172275875,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.2331,
|
|
"mean_token_accuracy": 0.09917902275919914,
|
|
"num_tokens": 1545401.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.569608402252197,
|
|
"epoch": 0.09072285630670178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.1165,
|
|
"mean_token_accuracy": 0.09994321763515472,
|
|
"num_tokens": 1561621.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.473207664489745,
|
|
"epoch": 0.09169837089064481,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.2154,
|
|
"mean_token_accuracy": 0.0949926421046257,
|
|
"num_tokens": 1579225.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.467814540863037,
|
|
"epoch": 0.09267388547458784,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.1298,
|
|
"mean_token_accuracy": 0.09709430187940597,
|
|
"num_tokens": 1596125.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.408341693878175,
|
|
"epoch": 0.09364940005853087,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.239,
|
|
"mean_token_accuracy": 0.0889017254114151,
|
|
"num_tokens": 1613451.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.418604183197022,
|
|
"epoch": 0.09462491464247391,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.1839,
|
|
"mean_token_accuracy": 0.09638622403144836,
|
|
"num_tokens": 1630929.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.463642883300782,
|
|
"epoch": 0.09560042922641694,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.1892,
|
|
"mean_token_accuracy": 0.09477911740541459,
|
|
"num_tokens": 1647379.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.404396533966064,
|
|
"epoch": 0.09657594381035997,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.972,
|
|
"mean_token_accuracy": 0.11155017167329788,
|
|
"num_tokens": 1664415.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.328124237060546,
|
|
"epoch": 0.097551458394303,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 8.0243,
|
|
"mean_token_accuracy": 0.10308937430381775,
|
|
"num_tokens": 1679556.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.29896330833435,
|
|
"epoch": 0.09852697297824603,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9748,
|
|
"mean_token_accuracy": 0.11530740708112716,
|
|
"num_tokens": 1696727.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.333364772796632,
|
|
"epoch": 0.09950248756218906,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 8.0291,
|
|
"mean_token_accuracy": 0.10489359647035598,
|
|
"num_tokens": 1713918.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.449989509582519,
|
|
"epoch": 0.10047800214613209,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 8.1751,
|
|
"mean_token_accuracy": 0.09076813086867333,
|
|
"num_tokens": 1729581.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.419713687896728,
|
|
"epoch": 0.10145351673007512,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 8.0506,
|
|
"mean_token_accuracy": 0.09246140196919442,
|
|
"num_tokens": 1746367.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.433179187774659,
|
|
"epoch": 0.10242903131401815,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 8.1352,
|
|
"mean_token_accuracy": 0.09465765729546546,
|
|
"num_tokens": 1765009.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.420462512969971,
|
|
"epoch": 0.10340454589796118,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 8.1402,
|
|
"mean_token_accuracy": 0.09433027505874633,
|
|
"num_tokens": 1782113.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.339337825775146,
|
|
"epoch": 0.10438006048190421,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 8.0357,
|
|
"mean_token_accuracy": 0.10014683678746224,
|
|
"num_tokens": 1798301.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.32430820465088,
|
|
"epoch": 0.10535557506584724,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 8.0744,
|
|
"mean_token_accuracy": 0.09614607095718383,
|
|
"num_tokens": 1815059.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.255896091461182,
|
|
"epoch": 0.10633108964979027,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.9775,
|
|
"mean_token_accuracy": 0.10225188136100768,
|
|
"num_tokens": 1831699.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.343412113189697,
|
|
"epoch": 0.1073066042337333,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 8.0561,
|
|
"mean_token_accuracy": 0.09823954552412033,
|
|
"num_tokens": 1847993.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.432454490661621,
|
|
"epoch": 0.10828211881767633,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9062,
|
|
"mean_token_accuracy": 0.11012097150087356,
|
|
"num_tokens": 1864753.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.282480907440185,
|
|
"epoch": 0.10925763340161936,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 8.06,
|
|
"mean_token_accuracy": 0.1011503592133522,
|
|
"num_tokens": 1882154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.209473037719727,
|
|
"epoch": 0.11023314798556239,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.9362,
|
|
"mean_token_accuracy": 0.106295907497406,
|
|
"num_tokens": 1899683.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.26357274055481,
|
|
"epoch": 0.11120866256950541,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.9382,
|
|
"mean_token_accuracy": 0.10315194055438041,
|
|
"num_tokens": 1918489.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.3274995803833,
|
|
"epoch": 0.11218417715344844,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9968,
|
|
"mean_token_accuracy": 0.09892351552844048,
|
|
"num_tokens": 1935609.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.20813455581665,
|
|
"epoch": 0.11315969173739147,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 8.0239,
|
|
"mean_token_accuracy": 0.09603053033351898,
|
|
"num_tokens": 1951311.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.337920379638671,
|
|
"epoch": 0.1141352063213345,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 8.0831,
|
|
"mean_token_accuracy": 0.0927000992000103,
|
|
"num_tokens": 1968908.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.290319728851319,
|
|
"epoch": 0.11511072090527753,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.9801,
|
|
"mean_token_accuracy": 0.09734337329864502,
|
|
"num_tokens": 1983807.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.349214029312133,
|
|
"epoch": 0.11608623548922056,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 8.0318,
|
|
"mean_token_accuracy": 0.09839608669281005,
|
|
"num_tokens": 2000219.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.2715407371521,
|
|
"epoch": 0.11706175007316359,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.9594,
|
|
"mean_token_accuracy": 0.1046835497021675,
|
|
"num_tokens": 2017360.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.223270702362061,
|
|
"epoch": 0.11803726465710662,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.9477,
|
|
"mean_token_accuracy": 0.10477901026606559,
|
|
"num_tokens": 2034953.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.252518939971925,
|
|
"epoch": 0.11901277924104965,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.921,
|
|
"mean_token_accuracy": 0.10113409459590912,
|
|
"num_tokens": 2051551.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.254705429077148,
|
|
"epoch": 0.11998829382499268,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.9086,
|
|
"mean_token_accuracy": 0.10554389208555222,
|
|
"num_tokens": 2067188.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.175243711471557,
|
|
"epoch": 0.12096380840893571,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.9407,
|
|
"mean_token_accuracy": 0.10011097714304924,
|
|
"num_tokens": 2082569.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.322654724121094,
|
|
"epoch": 0.12193932299287874,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 8.0026,
|
|
"mean_token_accuracy": 0.09715932160615921,
|
|
"num_tokens": 2099446.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.246174478530884,
|
|
"epoch": 0.12291483757682177,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.9585,
|
|
"mean_token_accuracy": 0.09914746657013893,
|
|
"num_tokens": 2116526.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.269398927688599,
|
|
"epoch": 0.1238903521607648,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.999,
|
|
"mean_token_accuracy": 0.09154902622103692,
|
|
"num_tokens": 2134624.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.124698829650878,
|
|
"epoch": 0.12486586674470783,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.8396,
|
|
"mean_token_accuracy": 0.10734372511506081,
|
|
"num_tokens": 2152872.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.231671237945557,
|
|
"epoch": 0.12584138132865086,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.9856,
|
|
"mean_token_accuracy": 0.09579248055815696,
|
|
"num_tokens": 2169480.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.27645206451416,
|
|
"epoch": 0.1268168959125939,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.8324,
|
|
"mean_token_accuracy": 0.10887673124670982,
|
|
"num_tokens": 2186054.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.135476446151733,
|
|
"epoch": 0.12779241049653692,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.9707,
|
|
"mean_token_accuracy": 0.095298021286726,
|
|
"num_tokens": 2202320.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.142270755767822,
|
|
"epoch": 0.12876792508047996,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.8372,
|
|
"mean_token_accuracy": 0.1047438345849514,
|
|
"num_tokens": 2218817.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.260239505767823,
|
|
"epoch": 0.12974343966442298,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.9913,
|
|
"mean_token_accuracy": 0.09537639319896699,
|
|
"num_tokens": 2234845.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.212341785430908,
|
|
"epoch": 0.13071895424836602,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.9264,
|
|
"mean_token_accuracy": 0.09926454424858093,
|
|
"num_tokens": 2251904.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.084620904922485,
|
|
"epoch": 0.13169446883230904,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.9275,
|
|
"mean_token_accuracy": 0.09966224431991577,
|
|
"num_tokens": 2268343.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.316670322418213,
|
|
"epoch": 0.13266998341625208,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.9125,
|
|
"mean_token_accuracy": 0.1013021558523178,
|
|
"num_tokens": 2284524.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.03240876197815,
|
|
"epoch": 0.1336454980001951,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.8445,
|
|
"mean_token_accuracy": 0.1054447665810585,
|
|
"num_tokens": 2302350.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.19091992378235,
|
|
"epoch": 0.13462101258413814,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.9228,
|
|
"mean_token_accuracy": 0.09539637044072151,
|
|
"num_tokens": 2319953.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 8.203334712982178,
|
|
"epoch": 0.13559652716808115,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.8172,
|
|
"mean_token_accuracy": 0.1024633839726448,
|
|
"num_tokens": 2336698.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.045712327957153,
|
|
"epoch": 0.1365720417520242,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7726,
|
|
"mean_token_accuracy": 0.10965236350893974,
|
|
"num_tokens": 2353986.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.118633937835693,
|
|
"epoch": 0.1375475563359672,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.8241,
|
|
"mean_token_accuracy": 0.10513764396309852,
|
|
"num_tokens": 2370320.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 8.149739933013915,
|
|
"epoch": 0.13852307091991026,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8644,
|
|
"mean_token_accuracy": 0.098866156488657,
|
|
"num_tokens": 2385453.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 8.12429337501526,
|
|
"epoch": 0.13949858550385327,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.8549,
|
|
"mean_token_accuracy": 0.09939151853322983,
|
|
"num_tokens": 2401545.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.115737628936767,
|
|
"epoch": 0.14047410008779632,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.8504,
|
|
"mean_token_accuracy": 0.10048539191484451,
|
|
"num_tokens": 2418414.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 8.118222570419311,
|
|
"epoch": 0.14144961467173933,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.9118,
|
|
"mean_token_accuracy": 0.10191983357071877,
|
|
"num_tokens": 2434973.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 8.17931604385376,
|
|
"epoch": 0.14242512925568238,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.8881,
|
|
"mean_token_accuracy": 0.09516675621271134,
|
|
"num_tokens": 2451053.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 8.09225902557373,
|
|
"epoch": 0.1434006438396254,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.8626,
|
|
"mean_token_accuracy": 0.09528173580765724,
|
|
"num_tokens": 2467280.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 8.088089656829833,
|
|
"epoch": 0.14437615842356843,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7803,
|
|
"mean_token_accuracy": 0.1092884823679924,
|
|
"num_tokens": 2484588.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 8.081604194641113,
|
|
"epoch": 0.14535167300751145,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.7932,
|
|
"mean_token_accuracy": 0.10359530225396156,
|
|
"num_tokens": 2501200.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 8.028561162948609,
|
|
"epoch": 0.1463271875914545,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7428,
|
|
"mean_token_accuracy": 0.11165913268923759,
|
|
"num_tokens": 2517199.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.071641397476196,
|
|
"epoch": 0.1473027021753975,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.8168,
|
|
"mean_token_accuracy": 0.09799458160996437,
|
|
"num_tokens": 2534824.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.968346309661865,
|
|
"epoch": 0.14827821675934055,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6936,
|
|
"mean_token_accuracy": 0.10773247331380845,
|
|
"num_tokens": 2550842.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 8.000276184082031,
|
|
"epoch": 0.14925373134328357,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6604,
|
|
"mean_token_accuracy": 0.10792932882905007,
|
|
"num_tokens": 2567878.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 8.091751098632812,
|
|
"epoch": 0.1502292459272266,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.7485,
|
|
"mean_token_accuracy": 0.10320784673094749,
|
|
"num_tokens": 2584148.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 8.008577871322633,
|
|
"epoch": 0.15120476051116966,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.7503,
|
|
"mean_token_accuracy": 0.10530681982636451,
|
|
"num_tokens": 2600067.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 8.026472663879394,
|
|
"epoch": 0.15218027509511267,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.7366,
|
|
"mean_token_accuracy": 0.09601326063275337,
|
|
"num_tokens": 2615746.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 8.045800304412841,
|
|
"epoch": 0.15315578967905571,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.7722,
|
|
"mean_token_accuracy": 0.10049763545393944,
|
|
"num_tokens": 2631641.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.96700086593628,
|
|
"epoch": 0.15413130426299873,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.7749,
|
|
"mean_token_accuracy": 0.10441548749804497,
|
|
"num_tokens": 2647015.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 8.038145112991334,
|
|
"epoch": 0.15510681884694177,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.687,
|
|
"mean_token_accuracy": 0.10431429743766785,
|
|
"num_tokens": 2664045.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.970753955841064,
|
|
"epoch": 0.1560823334308848,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.6672,
|
|
"mean_token_accuracy": 0.10528303459286689,
|
|
"num_tokens": 2680774.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 8.060584831237794,
|
|
"epoch": 0.15705784801482783,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.7653,
|
|
"mean_token_accuracy": 0.1024099849164486,
|
|
"num_tokens": 2697719.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.984896326065064,
|
|
"epoch": 0.15803336259877085,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.7287,
|
|
"mean_token_accuracy": 0.09548300430178643,
|
|
"num_tokens": 2714518.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.963473129272461,
|
|
"epoch": 0.1590088771827139,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.7758,
|
|
"mean_token_accuracy": 0.10250489190220832,
|
|
"num_tokens": 2729288.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 8.057403326034546,
|
|
"epoch": 0.1599843917666569,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.7589,
|
|
"mean_token_accuracy": 0.10371811836957931,
|
|
"num_tokens": 2746782.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.909966135025025,
|
|
"epoch": 0.16095990635059995,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.7918,
|
|
"mean_token_accuracy": 0.09433421418070793,
|
|
"num_tokens": 2763159.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.984905338287353,
|
|
"epoch": 0.16193542093454297,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.734,
|
|
"mean_token_accuracy": 0.10140182003378868,
|
|
"num_tokens": 2780338.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.930758285522461,
|
|
"epoch": 0.162910935518486,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.6397,
|
|
"mean_token_accuracy": 0.10677831694483757,
|
|
"num_tokens": 2796214.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 8.004572010040283,
|
|
"epoch": 0.16388645010242903,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.7556,
|
|
"mean_token_accuracy": 0.09714071974158286,
|
|
"num_tokens": 2812745.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.912738943099976,
|
|
"epoch": 0.16486196468637207,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.6233,
|
|
"mean_token_accuracy": 0.10320991724729538,
|
|
"num_tokens": 2828288.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.9848785400390625,
|
|
"epoch": 0.16583747927031509,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.7558,
|
|
"mean_token_accuracy": 0.0968838632106781,
|
|
"num_tokens": 2843618.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.911918306350708,
|
|
"epoch": 0.16681299385425813,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5565,
|
|
"mean_token_accuracy": 0.11261726319789886,
|
|
"num_tokens": 2860659.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.907874202728271,
|
|
"epoch": 0.16778850843820114,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.665,
|
|
"mean_token_accuracy": 0.10077697262167931,
|
|
"num_tokens": 2877482.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.96095380783081,
|
|
"epoch": 0.1687640230221442,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.7396,
|
|
"mean_token_accuracy": 0.09533020779490471,
|
|
"num_tokens": 2895868.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.936106157302857,
|
|
"epoch": 0.1697395376060872,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.7286,
|
|
"mean_token_accuracy": 0.1010709173977375,
|
|
"num_tokens": 2913103.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.909117889404297,
|
|
"epoch": 0.17071505219003025,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.6449,
|
|
"mean_token_accuracy": 0.1050123207271099,
|
|
"num_tokens": 2928291.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.895384120941162,
|
|
"epoch": 0.17169056677397326,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.6968,
|
|
"mean_token_accuracy": 0.10348322838544846,
|
|
"num_tokens": 2945115.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.935518503189087,
|
|
"epoch": 0.1726660813579163,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5932,
|
|
"mean_token_accuracy": 0.10359113737940788,
|
|
"num_tokens": 2962869.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.885776710510254,
|
|
"epoch": 0.17364159594185932,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.7148,
|
|
"mean_token_accuracy": 0.09867775589227676,
|
|
"num_tokens": 2979298.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.9616724967956545,
|
|
"epoch": 0.17461711052580237,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.6235,
|
|
"mean_token_accuracy": 0.10697314292192459,
|
|
"num_tokens": 2996146.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.861116552352906,
|
|
"epoch": 0.17559262510974538,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.6919,
|
|
"mean_token_accuracy": 0.10271590203046799,
|
|
"num_tokens": 3013075.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.901067304611206,
|
|
"epoch": 0.17656813969368843,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.6489,
|
|
"mean_token_accuracy": 0.10057242289185524,
|
|
"num_tokens": 3029494.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.8369715213775635,
|
|
"epoch": 0.17754365427763144,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.6549,
|
|
"mean_token_accuracy": 0.10227547287940979,
|
|
"num_tokens": 3046333.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.902799129486084,
|
|
"epoch": 0.17851916886157448,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.6405,
|
|
"mean_token_accuracy": 0.1027099683880806,
|
|
"num_tokens": 3062714.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.915635061264038,
|
|
"epoch": 0.1794946834455175,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.6243,
|
|
"mean_token_accuracy": 0.10555409193038941,
|
|
"num_tokens": 3080711.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.877739381790161,
|
|
"epoch": 0.18047019802946054,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.7127,
|
|
"mean_token_accuracy": 0.09761228263378144,
|
|
"num_tokens": 3096885.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.808351802825928,
|
|
"epoch": 0.18144571261340356,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.607,
|
|
"mean_token_accuracy": 0.1075299970805645,
|
|
"num_tokens": 3114536.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.82890100479126,
|
|
"epoch": 0.1824212271973466,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5659,
|
|
"mean_token_accuracy": 0.10890542268753052,
|
|
"num_tokens": 3131921.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.862617540359497,
|
|
"epoch": 0.18339674178128962,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.7037,
|
|
"mean_token_accuracy": 0.09914149567484856,
|
|
"num_tokens": 3148064.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.748774862289428,
|
|
"epoch": 0.18437225636523266,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.527,
|
|
"mean_token_accuracy": 0.10934840366244317,
|
|
"num_tokens": 3165822.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.820528364181518,
|
|
"epoch": 0.18534777094917568,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.6245,
|
|
"mean_token_accuracy": 0.10755226612091065,
|
|
"num_tokens": 3183286.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.906544256210327,
|
|
"epoch": 0.18632328553311872,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.6609,
|
|
"mean_token_accuracy": 0.09516188427805901,
|
|
"num_tokens": 3198522.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.809338474273682,
|
|
"epoch": 0.18729880011706174,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.6494,
|
|
"mean_token_accuracy": 0.10158563554286956,
|
|
"num_tokens": 3218422.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.857818698883056,
|
|
"epoch": 0.18827431470100478,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.694,
|
|
"mean_token_accuracy": 0.09597188532352448,
|
|
"num_tokens": 3236407.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.813616561889648,
|
|
"epoch": 0.18924982928494782,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.6223,
|
|
"mean_token_accuracy": 0.10206324681639671,
|
|
"num_tokens": 3252740.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.8036384105682375,
|
|
"epoch": 0.19022534386889084,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5892,
|
|
"mean_token_accuracy": 0.10098864287137985,
|
|
"num_tokens": 3269874.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.835381650924683,
|
|
"epoch": 0.19120085845283388,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.6163,
|
|
"mean_token_accuracy": 0.09953035637736321,
|
|
"num_tokens": 3286515.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.6930643081665036,
|
|
"epoch": 0.1921763730367769,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4293,
|
|
"mean_token_accuracy": 0.11725058928132057,
|
|
"num_tokens": 3303700.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.785474395751953,
|
|
"epoch": 0.19315188762071994,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5881,
|
|
"mean_token_accuracy": 0.10581063255667686,
|
|
"num_tokens": 3319768.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.838024091720581,
|
|
"epoch": 0.19412740220466296,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4995,
|
|
"mean_token_accuracy": 0.10780946016311646,
|
|
"num_tokens": 3336555.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.650297355651856,
|
|
"epoch": 0.195102916788606,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.5032,
|
|
"mean_token_accuracy": 0.10967954769730567,
|
|
"num_tokens": 3353273.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.755049371719361,
|
|
"epoch": 0.19607843137254902,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999999929644166,
|
|
"loss": 7.4093,
|
|
"mean_token_accuracy": 0.1141424149274826,
|
|
"num_tokens": 3369408.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.701835823059082,
|
|
"epoch": 0.19705394595649206,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999999643823598,
|
|
"loss": 7.4702,
|
|
"mean_token_accuracy": 0.10950530394911766,
|
|
"num_tokens": 3385834.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.682157421112061,
|
|
"epoch": 0.19802946054043508,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999999138141084,
|
|
"loss": 7.5624,
|
|
"mean_token_accuracy": 0.10582574605941772,
|
|
"num_tokens": 3401407.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.864960336685181,
|
|
"epoch": 0.19900497512437812,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004999998412596673,
|
|
"loss": 7.5728,
|
|
"mean_token_accuracy": 0.0993146575987339,
|
|
"num_tokens": 3418256.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.658085298538208,
|
|
"epoch": 0.19998048970832114,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999997467190435,
|
|
"loss": 7.5535,
|
|
"mean_token_accuracy": 0.10653921887278557,
|
|
"num_tokens": 3433932.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.765207004547119,
|
|
"epoch": 0.20095600429226418,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999996301922465,
|
|
"loss": 7.5996,
|
|
"mean_token_accuracy": 0.09861309975385665,
|
|
"num_tokens": 3451723.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.730470132827759,
|
|
"epoch": 0.2019315188762072,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999994916792876,
|
|
"loss": 7.5662,
|
|
"mean_token_accuracy": 0.1022422194480896,
|
|
"num_tokens": 3469170.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.746661567687989,
|
|
"epoch": 0.20290703346015024,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999993311801802,
|
|
"loss": 7.5806,
|
|
"mean_token_accuracy": 0.10085131973028183,
|
|
"num_tokens": 3485282.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.7354357719421385,
|
|
"epoch": 0.20388254804409325,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.00049999914869494,
|
|
"loss": 7.5625,
|
|
"mean_token_accuracy": 0.09602305442094802,
|
|
"num_tokens": 3502206.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.754732227325439,
|
|
"epoch": 0.2048580626280363,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000499998944223585,
|
|
"loss": 7.4983,
|
|
"mean_token_accuracy": 0.10457974448800086,
|
|
"num_tokens": 3520384.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.653174543380738,
|
|
"epoch": 0.2058335772119793,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004999987177661351,
|
|
"loss": 7.5054,
|
|
"mean_token_accuracy": 0.10512347519397736,
|
|
"num_tokens": 3536314.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.6191648006439205,
|
|
"epoch": 0.20680909179592236,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999984693226124,
|
|
"loss": 7.4757,
|
|
"mean_token_accuracy": 0.11180041655898094,
|
|
"num_tokens": 3551510.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.7276811599731445,
|
|
"epoch": 0.20778460637986537,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999981988930411,
|
|
"loss": 7.4781,
|
|
"mean_token_accuracy": 0.10681704133749008,
|
|
"num_tokens": 3568515.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.692039155960083,
|
|
"epoch": 0.20876012096380842,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999979064774478,
|
|
"loss": 7.521,
|
|
"mean_token_accuracy": 0.1076972097158432,
|
|
"num_tokens": 3585183.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.666302013397217,
|
|
"epoch": 0.20973563554775143,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499997592075861,
|
|
"loss": 7.4576,
|
|
"mean_token_accuracy": 0.10820185318589211,
|
|
"num_tokens": 3601898.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.777315092086792,
|
|
"epoch": 0.21071115013169447,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999972556883113,
|
|
"loss": 7.5112,
|
|
"mean_token_accuracy": 0.10493268743157387,
|
|
"num_tokens": 3616699.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.650645208358765,
|
|
"epoch": 0.2116866647156375,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999968973148318,
|
|
"loss": 7.5255,
|
|
"mean_token_accuracy": 0.10378064140677452,
|
|
"num_tokens": 3634030.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.74101037979126,
|
|
"epoch": 0.21266217929958053,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004999965169554573,
|
|
"loss": 7.5187,
|
|
"mean_token_accuracy": 0.10108392685651779,
|
|
"num_tokens": 3650495.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.681738424301147,
|
|
"epoch": 0.21363769388352355,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999961146102251,
|
|
"loss": 7.485,
|
|
"mean_token_accuracy": 0.10594045221805573,
|
|
"num_tokens": 3666926.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.639343309402466,
|
|
"epoch": 0.2146132084674666,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004999956902791746,
|
|
"loss": 7.4543,
|
|
"mean_token_accuracy": 0.11311954259872437,
|
|
"num_tokens": 3683061.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.593081951141357,
|
|
"epoch": 0.2155887230514096,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004999952439623469,
|
|
"loss": 7.5121,
|
|
"mean_token_accuracy": 0.10505839735269547,
|
|
"num_tokens": 3703013.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.679114913940429,
|
|
"epoch": 0.21656423763535265,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000499994775659786,
|
|
"loss": 7.3601,
|
|
"mean_token_accuracy": 0.11591144353151321,
|
|
"num_tokens": 3719885.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.631566095352173,
|
|
"epoch": 0.21753975221929567,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999942853715375,
|
|
"loss": 7.4971,
|
|
"mean_token_accuracy": 0.10333904549479485,
|
|
"num_tokens": 3735122.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.694211006164551,
|
|
"epoch": 0.2185152668032387,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004999937730976494,
|
|
"loss": 7.4806,
|
|
"mean_token_accuracy": 0.10258661657571792,
|
|
"num_tokens": 3751627.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.595678424835205,
|
|
"epoch": 0.21949078138718173,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004999932388381716,
|
|
"loss": 7.4929,
|
|
"mean_token_accuracy": 0.101930870115757,
|
|
"num_tokens": 3768165.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.726031064987183,
|
|
"epoch": 0.22046629597112477,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004999926825931564,
|
|
"loss": 7.5163,
|
|
"mean_token_accuracy": 0.10487755760550499,
|
|
"num_tokens": 3784245.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.5429949283599855,
|
|
"epoch": 0.2214418105550678,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004999921043626582,
|
|
"loss": 7.2742,
|
|
"mean_token_accuracy": 0.11879447922110557,
|
|
"num_tokens": 3802057.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.616366910934448,
|
|
"epoch": 0.22241732513901083,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999915041467333,
|
|
"loss": 7.4538,
|
|
"mean_token_accuracy": 0.10403451472520828,
|
|
"num_tokens": 3817104.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.623820781707764,
|
|
"epoch": 0.22339283972295385,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0004999908819454407,
|
|
"loss": 7.4557,
|
|
"mean_token_accuracy": 0.10573292225599289,
|
|
"num_tokens": 3833037.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.591778421401978,
|
|
"epoch": 0.2243683543068969,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004999902377588408,
|
|
"loss": 7.3454,
|
|
"mean_token_accuracy": 0.1075218863785267,
|
|
"num_tokens": 3849634.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.620333814620972,
|
|
"epoch": 0.2253438688908399,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000499989571586997,
|
|
"loss": 7.5125,
|
|
"mean_token_accuracy": 0.10198941007256508,
|
|
"num_tokens": 3865949.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.529477119445801,
|
|
"epoch": 0.22631938347478295,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004999888834299739,
|
|
"loss": 7.3796,
|
|
"mean_token_accuracy": 0.11067740991711617,
|
|
"num_tokens": 3884646.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.658005952835083,
|
|
"epoch": 0.227294898058726,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004999881732878391,
|
|
"loss": 7.4808,
|
|
"mean_token_accuracy": 0.10823895409703255,
|
|
"num_tokens": 3900740.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.605532455444336,
|
|
"epoch": 0.228270412642669,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999874411606618,
|
|
"loss": 7.4248,
|
|
"mean_token_accuracy": 0.10825219377875328,
|
|
"num_tokens": 3918619.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.614225101470947,
|
|
"epoch": 0.22924592722661205,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004999866870485137,
|
|
"loss": 7.3986,
|
|
"mean_token_accuracy": 0.10738857686519623,
|
|
"num_tokens": 3938161.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.632542467117309,
|
|
"epoch": 0.23022144181055507,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999859109514684,
|
|
"loss": 7.4496,
|
|
"mean_token_accuracy": 0.10592144280672074,
|
|
"num_tokens": 3953582.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.637817001342773,
|
|
"epoch": 0.2311969563944981,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004999851128696017,
|
|
"loss": 7.4297,
|
|
"mean_token_accuracy": 0.09909034296870231,
|
|
"num_tokens": 3969747.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.631790590286255,
|
|
"epoch": 0.23217247097844113,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004999842928029917,
|
|
"loss": 7.4596,
|
|
"mean_token_accuracy": 0.10731764733791352,
|
|
"num_tokens": 3984646.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.594443273544312,
|
|
"epoch": 0.23314798556238417,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004999834507517184,
|
|
"loss": 7.4381,
|
|
"mean_token_accuracy": 0.10461552739143372,
|
|
"num_tokens": 4003108.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.619934177398681,
|
|
"epoch": 0.23412350014632718,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999825867158642,
|
|
"loss": 7.5028,
|
|
"mean_token_accuracy": 0.10035586953163148,
|
|
"num_tokens": 4019944.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.667219352722168,
|
|
"epoch": 0.23509901473027023,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004999817006955134,
|
|
"loss": 7.3921,
|
|
"mean_token_accuracy": 0.10938728377223014,
|
|
"num_tokens": 4037617.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.553360652923584,
|
|
"epoch": 0.23607452931421324,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004999807926907527,
|
|
"loss": 7.3945,
|
|
"mean_token_accuracy": 0.10081385970115661,
|
|
"num_tokens": 4053483.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.64829454421997,
|
|
"epoch": 0.2370500438981563,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999798627016707,
|
|
"loss": 7.4587,
|
|
"mean_token_accuracy": 0.10884808897972106,
|
|
"num_tokens": 4070165.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.619628667831421,
|
|
"epoch": 0.2380255584820993,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999789107283586,
|
|
"loss": 7.4342,
|
|
"mean_token_accuracy": 0.1030805915594101,
|
|
"num_tokens": 4086208.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.502168989181518,
|
|
"epoch": 0.23900107306604235,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.000499977936770909,
|
|
"loss": 7.4058,
|
|
"mean_token_accuracy": 0.10911756455898285,
|
|
"num_tokens": 4101790.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.568706035614014,
|
|
"epoch": 0.23997658764998536,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999769408294173,
|
|
"loss": 7.3363,
|
|
"mean_token_accuracy": 0.11626880168914795,
|
|
"num_tokens": 4117633.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.568750286102295,
|
|
"epoch": 0.2409521022339284,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999759229039807,
|
|
"loss": 7.3873,
|
|
"mean_token_accuracy": 0.10927161276340484,
|
|
"num_tokens": 4132634.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.5792591094970705,
|
|
"epoch": 0.24192761681787142,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0004999748829946988,
|
|
"loss": 7.442,
|
|
"mean_token_accuracy": 0.10532458499073982,
|
|
"num_tokens": 4150702.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.524825191497802,
|
|
"epoch": 0.24290313140181446,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999738211016732,
|
|
"loss": 7.3054,
|
|
"mean_token_accuracy": 0.12065900191664695,
|
|
"num_tokens": 4168483.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.430249309539795,
|
|
"epoch": 0.24387864598575748,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004999727372250076,
|
|
"loss": 7.2814,
|
|
"mean_token_accuracy": 0.11273804232478142,
|
|
"num_tokens": 4186199.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.560726976394653,
|
|
"epoch": 0.24485416056970052,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999716313648079,
|
|
"loss": 7.4101,
|
|
"mean_token_accuracy": 0.10708601549267768,
|
|
"num_tokens": 4202303.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.622148323059082,
|
|
"epoch": 0.24582967515364354,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999705035211822,
|
|
"loss": 7.4547,
|
|
"mean_token_accuracy": 0.10375886857509613,
|
|
"num_tokens": 4218541.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.5890209674835205,
|
|
"epoch": 0.24680518973758658,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999693536942406,
|
|
"loss": 7.4123,
|
|
"mean_token_accuracy": 0.10459026619791985,
|
|
"num_tokens": 4234025.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.591500186920166,
|
|
"epoch": 0.2477807043215296,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999681818840957,
|
|
"loss": 7.4257,
|
|
"mean_token_accuracy": 0.11080710366368293,
|
|
"num_tokens": 4250798.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.561366319656372,
|
|
"epoch": 0.24875621890547264,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000499966988090862,
|
|
"loss": 7.4261,
|
|
"mean_token_accuracy": 0.10205279141664506,
|
|
"num_tokens": 4266453.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.638256502151489,
|
|
"epoch": 0.24973173348941566,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999657723146558,
|
|
"loss": 7.3818,
|
|
"mean_token_accuracy": 0.10573315620422363,
|
|
"num_tokens": 4281825.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.389687967300415,
|
|
"epoch": 0.2507072480733587,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004999645345555962,
|
|
"loss": 7.2205,
|
|
"mean_token_accuracy": 0.12211042940616608,
|
|
"num_tokens": 4298821.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.507760238647461,
|
|
"epoch": 0.2516827626573017,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999632748138041,
|
|
"loss": 7.267,
|
|
"mean_token_accuracy": 0.11187295690178871,
|
|
"num_tokens": 4314076.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.534629392623901,
|
|
"epoch": 0.25265827724124473,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004999619930894025,
|
|
"loss": 7.2723,
|
|
"mean_token_accuracy": 0.1124474972486496,
|
|
"num_tokens": 4330909.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.508888912200928,
|
|
"epoch": 0.2536337918251878,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004999606893825168,
|
|
"loss": 7.367,
|
|
"mean_token_accuracy": 0.10985746160149575,
|
|
"num_tokens": 4346927.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.526936769485474,
|
|
"epoch": 0.2546093064091308,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0004999593636932742,
|
|
"loss": 7.4001,
|
|
"mean_token_accuracy": 0.10443056225776673,
|
|
"num_tokens": 4364956.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.524621152877808,
|
|
"epoch": 0.25558482099307384,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0004999580160218044,
|
|
"loss": 7.3355,
|
|
"mean_token_accuracy": 0.10670254677534104,
|
|
"num_tokens": 4383014.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.467378330230713,
|
|
"epoch": 0.25656033557701685,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0004999566463682391,
|
|
"loss": 7.4095,
|
|
"mean_token_accuracy": 0.10816679820418358,
|
|
"num_tokens": 4401050.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.533441591262817,
|
|
"epoch": 0.2575358501609599,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004999552547327119,
|
|
"loss": 7.3048,
|
|
"mean_token_accuracy": 0.10800013542175294,
|
|
"num_tokens": 4415453.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.4499256134033205,
|
|
"epoch": 0.25851136474490294,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999538411153591,
|
|
"loss": 7.2945,
|
|
"mean_token_accuracy": 0.11506965532898902,
|
|
"num_tokens": 4433107.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.55745005607605,
|
|
"epoch": 0.25948687932884595,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999524055163187,
|
|
"loss": 7.373,
|
|
"mean_token_accuracy": 0.10276507884263993,
|
|
"num_tokens": 4449440.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.562828826904297,
|
|
"epoch": 0.26046239391278897,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000499950947935731,
|
|
"loss": 7.3544,
|
|
"mean_token_accuracy": 0.10752240270376205,
|
|
"num_tokens": 4465327.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.5065587043762205,
|
|
"epoch": 0.26143790849673204,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004999494683737384,
|
|
"loss": 7.3836,
|
|
"mean_token_accuracy": 0.10438340827822686,
|
|
"num_tokens": 4481518.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.607649135589599,
|
|
"epoch": 0.26241342308067506,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004999479668304854,
|
|
"loss": 7.4024,
|
|
"mean_token_accuracy": 0.10401526093482971,
|
|
"num_tokens": 4497632.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.285141897201538,
|
|
"epoch": 0.2633889376646181,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004999464433061188,
|
|
"loss": 7.1554,
|
|
"mean_token_accuracy": 0.12241300269961357,
|
|
"num_tokens": 4513790.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.519728755950927,
|
|
"epoch": 0.26436445224856114,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004999448978007876,
|
|
"loss": 7.2984,
|
|
"mean_token_accuracy": 0.11036163046956063,
|
|
"num_tokens": 4530802.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.546466922760009,
|
|
"epoch": 0.26533996683250416,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999433303146426,
|
|
"loss": 7.3356,
|
|
"mean_token_accuracy": 0.1046901948750019,
|
|
"num_tokens": 4546208.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.4623072147369385,
|
|
"epoch": 0.2663154814164472,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004999417408478371,
|
|
"loss": 7.3911,
|
|
"mean_token_accuracy": 0.10712278932332993,
|
|
"num_tokens": 4563753.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.459856939315796,
|
|
"epoch": 0.2672909960003902,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999401294005264,
|
|
"loss": 7.273,
|
|
"mean_token_accuracy": 0.11257142499089241,
|
|
"num_tokens": 4580274.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.477842092514038,
|
|
"epoch": 0.26826651058433326,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000499938495972868,
|
|
"loss": 7.3515,
|
|
"mean_token_accuracy": 0.10793646723031998,
|
|
"num_tokens": 4596085.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.486316585540772,
|
|
"epoch": 0.2692420251682763,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004999368405650214,
|
|
"loss": 7.381,
|
|
"mean_token_accuracy": 0.10322927758097648,
|
|
"num_tokens": 4613258.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.384283590316772,
|
|
"epoch": 0.2702175397522193,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004999351631771485,
|
|
"loss": 7.2098,
|
|
"mean_token_accuracy": 0.11609927490353585,
|
|
"num_tokens": 4630128.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.547839498519897,
|
|
"epoch": 0.2711930543361623,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004999334638094131,
|
|
"loss": 7.2417,
|
|
"mean_token_accuracy": 0.11260311901569367,
|
|
"num_tokens": 4646181.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.463831806182862,
|
|
"epoch": 0.2721685689201054,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999317424619812,
|
|
"loss": 7.299,
|
|
"mean_token_accuracy": 0.10586258545517921,
|
|
"num_tokens": 4661804.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.43352689743042,
|
|
"epoch": 0.2731440835040484,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004999299991350212,
|
|
"loss": 7.2196,
|
|
"mean_token_accuracy": 0.11394756436347961,
|
|
"num_tokens": 4678163.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.499938440322876,
|
|
"epoch": 0.2741195980879914,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999282338287034,
|
|
"loss": 7.2102,
|
|
"mean_token_accuracy": 0.11337384283542633,
|
|
"num_tokens": 4695469.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.501874732971191,
|
|
"epoch": 0.2750951126719344,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999264465432001,
|
|
"loss": 7.3461,
|
|
"mean_token_accuracy": 0.11030107736587524,
|
|
"num_tokens": 4711599.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.478948163986206,
|
|
"epoch": 0.2760706272558775,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999246372786862,
|
|
"loss": 7.3544,
|
|
"mean_token_accuracy": 0.10497529283165932,
|
|
"num_tokens": 4728381.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.4203221797943115,
|
|
"epoch": 0.2770461418398205,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004999228060353385,
|
|
"loss": 7.3363,
|
|
"mean_token_accuracy": 0.10868120938539505,
|
|
"num_tokens": 4745348.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.5548547267913815,
|
|
"epoch": 0.27802165642376353,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999209528133357,
|
|
"loss": 7.3448,
|
|
"mean_token_accuracy": 0.10266721099615098,
|
|
"num_tokens": 4763011.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.4765321731567385,
|
|
"epoch": 0.27899717100770655,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000499919077612859,
|
|
"loss": 7.2882,
|
|
"mean_token_accuracy": 0.10844862014055252,
|
|
"num_tokens": 4779462.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.293780994415283,
|
|
"epoch": 0.2799726855916496,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999171804340917,
|
|
"loss": 7.2313,
|
|
"mean_token_accuracy": 0.11544499918818474,
|
|
"num_tokens": 4796844.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.474391031265259,
|
|
"epoch": 0.28094820017559263,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004999152612772191,
|
|
"loss": 7.2579,
|
|
"mean_token_accuracy": 0.11364135891199112,
|
|
"num_tokens": 4812851.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.348306369781494,
|
|
"epoch": 0.28192371475953565,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999133201424288,
|
|
"loss": 7.2109,
|
|
"mean_token_accuracy": 0.1137004129588604,
|
|
"num_tokens": 4830144.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.589390277862549,
|
|
"epoch": 0.28289922934347866,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004999113570299105,
|
|
"loss": 7.3162,
|
|
"mean_token_accuracy": 0.10344778746366501,
|
|
"num_tokens": 4846981.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.439992523193359,
|
|
"epoch": 0.28387474392742174,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004999093719398561,
|
|
"loss": 7.4507,
|
|
"mean_token_accuracy": 0.0993297427892685,
|
|
"num_tokens": 4864404.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.506286716461181,
|
|
"epoch": 0.28485025851136475,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999073648724592,
|
|
"loss": 7.3368,
|
|
"mean_token_accuracy": 0.10415377318859101,
|
|
"num_tokens": 4882004.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.455475807189941,
|
|
"epoch": 0.28582577309530777,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999053358279164,
|
|
"loss": 7.2824,
|
|
"mean_token_accuracy": 0.10441663935780525,
|
|
"num_tokens": 4899444.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.507974910736084,
|
|
"epoch": 0.2868012876792508,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004999032848064256,
|
|
"loss": 7.2713,
|
|
"mean_token_accuracy": 0.105850101262331,
|
|
"num_tokens": 4917214.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.377426624298096,
|
|
"epoch": 0.28777680226319385,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999012118081875,
|
|
"loss": 7.3534,
|
|
"mean_token_accuracy": 0.1025879755616188,
|
|
"num_tokens": 4934534.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.548637533187867,
|
|
"epoch": 0.28875231684713687,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004998991168334045,
|
|
"loss": 7.2657,
|
|
"mean_token_accuracy": 0.10621999204158783,
|
|
"num_tokens": 4950740.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.444481134414673,
|
|
"epoch": 0.2897278314310799,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998969998822814,
|
|
"loss": 7.3529,
|
|
"mean_token_accuracy": 0.10245063826441765,
|
|
"num_tokens": 4966720.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.47595624923706,
|
|
"epoch": 0.2907033460150229,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004998948609550249,
|
|
"loss": 7.245,
|
|
"mean_token_accuracy": 0.10455562472343445,
|
|
"num_tokens": 4982195.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.48439302444458,
|
|
"epoch": 0.29167886059896597,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998927000518442,
|
|
"loss": 7.3514,
|
|
"mean_token_accuracy": 0.10247311592102051,
|
|
"num_tokens": 4997487.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.406205940246582,
|
|
"epoch": 0.292654375182909,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998905171729505,
|
|
"loss": 7.2324,
|
|
"mean_token_accuracy": 0.10920758843421936,
|
|
"num_tokens": 5012913.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.351927471160889,
|
|
"epoch": 0.293629889766852,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004998883123185568,
|
|
"loss": 7.1937,
|
|
"mean_token_accuracy": 0.12020321637392044,
|
|
"num_tokens": 5031462.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.460276126861572,
|
|
"epoch": 0.294605404350795,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998860854888788,
|
|
"loss": 7.1911,
|
|
"mean_token_accuracy": 0.11498533636331558,
|
|
"num_tokens": 5047354.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.336567544937134,
|
|
"epoch": 0.2955809189347381,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.0004998838366841341,
|
|
"loss": 7.2655,
|
|
"mean_token_accuracy": 0.11058657392859458,
|
|
"num_tokens": 5065268.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.4529026508331295,
|
|
"epoch": 0.2965564335186811,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004998815659045423,
|
|
"loss": 7.2507,
|
|
"mean_token_accuracy": 0.11000868156552315,
|
|
"num_tokens": 5081066.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.428559732437134,
|
|
"epoch": 0.2975319481026241,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004998792731503254,
|
|
"loss": 7.284,
|
|
"mean_token_accuracy": 0.1131756454706192,
|
|
"num_tokens": 5099440.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.379742431640625,
|
|
"epoch": 0.29850746268656714,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004998769584217074,
|
|
"loss": 7.2445,
|
|
"mean_token_accuracy": 0.11773268282413482,
|
|
"num_tokens": 5116573.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.285349035263062,
|
|
"epoch": 0.2994829772705102,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004998746217189145,
|
|
"loss": 7.0154,
|
|
"mean_token_accuracy": 0.12612679973244667,
|
|
"num_tokens": 5132920.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.485317373275757,
|
|
"epoch": 0.3004584918544532,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000499872263042175,
|
|
"loss": 7.2709,
|
|
"mean_token_accuracy": 0.10634980946779252,
|
|
"num_tokens": 5150681.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.2799914360046385,
|
|
"epoch": 0.30143400643839624,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004998698823917196,
|
|
"loss": 7.0609,
|
|
"mean_token_accuracy": 0.12634149193763733,
|
|
"num_tokens": 5167473.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.412469148635864,
|
|
"epoch": 0.3024095210223393,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0004998674797677805,
|
|
"loss": 7.329,
|
|
"mean_token_accuracy": 0.10400748774409294,
|
|
"num_tokens": 5183847.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.394794702529907,
|
|
"epoch": 0.3033850356062823,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998650551705929,
|
|
"loss": 7.2701,
|
|
"mean_token_accuracy": 0.1091896690428257,
|
|
"num_tokens": 5200019.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.368698263168335,
|
|
"epoch": 0.30436055019022534,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998626086003934,
|
|
"loss": 7.1482,
|
|
"mean_token_accuracy": 0.11328554674983024,
|
|
"num_tokens": 5217245.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.3528977394104,
|
|
"epoch": 0.30533606477416836,
|
|
"grad_norm": 0.66015625,
|
|
"learning_rate": 0.0004998601400574213,
|
|
"loss": 7.047,
|
|
"mean_token_accuracy": 0.11916083097457886,
|
|
"num_tokens": 5234822.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.351659297943115,
|
|
"epoch": 0.30631157935811143,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998576495419177,
|
|
"loss": 7.1878,
|
|
"mean_token_accuracy": 0.10867269039154052,
|
|
"num_tokens": 5251282.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.320261383056641,
|
|
"epoch": 0.30728709394205445,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.000499855137054126,
|
|
"loss": 7.1493,
|
|
"mean_token_accuracy": 0.11374075636267662,
|
|
"num_tokens": 5269630.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.461973571777344,
|
|
"epoch": 0.30826260852599746,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.0004998526025942917,
|
|
"loss": 7.231,
|
|
"mean_token_accuracy": 0.10755549967288972,
|
|
"num_tokens": 5287106.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.371440410614014,
|
|
"epoch": 0.3092381231099405,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004998500461626625,
|
|
"loss": 7.265,
|
|
"mean_token_accuracy": 0.10544228777289391,
|
|
"num_tokens": 5303387.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.4729814529418945,
|
|
"epoch": 0.31021363769388355,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000499847467759488,
|
|
"loss": 7.2561,
|
|
"mean_token_accuracy": 0.10558799877762795,
|
|
"num_tokens": 5320482.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.344847202301025,
|
|
"epoch": 0.31118915227782656,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004998448673850205,
|
|
"loss": 7.2078,
|
|
"mean_token_accuracy": 0.1119214117527008,
|
|
"num_tokens": 5336420.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.33606595993042,
|
|
"epoch": 0.3121646668617696,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004998422450395139,
|
|
"loss": 7.1265,
|
|
"mean_token_accuracy": 0.1145975984632969,
|
|
"num_tokens": 5353464.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.326481723785401,
|
|
"epoch": 0.3131401814457126,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004998396007232244,
|
|
"loss": 7.2097,
|
|
"mean_token_accuracy": 0.1147161565721035,
|
|
"num_tokens": 5372607.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.383757877349853,
|
|
"epoch": 0.31411569602965567,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004998369344364105,
|
|
"loss": 7.2393,
|
|
"mean_token_accuracy": 0.1151178702712059,
|
|
"num_tokens": 5390134.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.347494220733642,
|
|
"epoch": 0.3150912106135987,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004998342461793328,
|
|
"loss": 7.1527,
|
|
"mean_token_accuracy": 0.11332457587122917,
|
|
"num_tokens": 5407083.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.2023365020751955,
|
|
"epoch": 0.3160667251975417,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004998315359522538,
|
|
"loss": 7.0363,
|
|
"mean_token_accuracy": 0.12925028875470163,
|
|
"num_tokens": 5424617.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.328793096542358,
|
|
"epoch": 0.3170422397814847,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004998288037554383,
|
|
"loss": 7.1895,
|
|
"mean_token_accuracy": 0.11991029530763626,
|
|
"num_tokens": 5440913.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.3388244152069095,
|
|
"epoch": 0.3180177543654278,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004998260495891535,
|
|
"loss": 7.2081,
|
|
"mean_token_accuracy": 0.11506202220916747,
|
|
"num_tokens": 5457157.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.34562315940857,
|
|
"epoch": 0.3189932689493708,
|
|
"grad_norm": 0.64453125,
|
|
"learning_rate": 0.0004998232734536685,
|
|
"loss": 7.1309,
|
|
"mean_token_accuracy": 0.11558853089809418,
|
|
"num_tokens": 5474520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.379280471801758,
|
|
"epoch": 0.3199687835333138,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998204753492544,
|
|
"loss": 7.1992,
|
|
"mean_token_accuracy": 0.11453557461500168,
|
|
"num_tokens": 5490599.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.365497016906739,
|
|
"epoch": 0.32094429811725683,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004998176552761848,
|
|
"loss": 7.2678,
|
|
"mean_token_accuracy": 0.10158305242657661,
|
|
"num_tokens": 5506714.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.373086643218994,
|
|
"epoch": 0.3219198127011999,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004998148132347352,
|
|
"loss": 7.221,
|
|
"mean_token_accuracy": 0.10766603276133538,
|
|
"num_tokens": 5522894.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.307815074920654,
|
|
"epoch": 0.3228953272851429,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004998119492251832,
|
|
"loss": 7.099,
|
|
"mean_token_accuracy": 0.11964272856712341,
|
|
"num_tokens": 5540054.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.383569002151489,
|
|
"epoch": 0.32387084186908593,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998090632478088,
|
|
"loss": 7.2552,
|
|
"mean_token_accuracy": 0.1003120444715023,
|
|
"num_tokens": 5556189.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.420545959472657,
|
|
"epoch": 0.32484635645302895,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000499806155302894,
|
|
"loss": 7.2685,
|
|
"mean_token_accuracy": 0.10375322699546814,
|
|
"num_tokens": 5572562.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.390063381195068,
|
|
"epoch": 0.325821871036972,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000499803225390723,
|
|
"loss": 7.2527,
|
|
"mean_token_accuracy": 0.10695536509156227,
|
|
"num_tokens": 5588992.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.4009096145629885,
|
|
"epoch": 0.32679738562091504,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004998002735115819,
|
|
"loss": 7.1757,
|
|
"mean_token_accuracy": 0.11011699885129929,
|
|
"num_tokens": 5604143.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.338191652297974,
|
|
"epoch": 0.32777290020485805,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004997972996657593,
|
|
"loss": 7.3027,
|
|
"mean_token_accuracy": 0.09990110099315644,
|
|
"num_tokens": 5620887.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.4250908374786375,
|
|
"epoch": 0.32874841478880107,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004997943038535459,
|
|
"loss": 7.2363,
|
|
"mean_token_accuracy": 0.11051089987158776,
|
|
"num_tokens": 5637156.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.357939052581787,
|
|
"epoch": 0.32972392937274414,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004997912860752341,
|
|
"loss": 7.1842,
|
|
"mean_token_accuracy": 0.10788781493902207,
|
|
"num_tokens": 5654024.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.352538108825684,
|
|
"epoch": 0.33069944395668716,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997882463311192,
|
|
"loss": 7.1334,
|
|
"mean_token_accuracy": 0.11298866793513299,
|
|
"num_tokens": 5671538.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.3202063083648685,
|
|
"epoch": 0.33167495854063017,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004997851846214979,
|
|
"loss": 7.1849,
|
|
"mean_token_accuracy": 0.11629349514842033,
|
|
"num_tokens": 5688788.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.302315759658813,
|
|
"epoch": 0.3326504731245732,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0004997821009466695,
|
|
"loss": 7.1253,
|
|
"mean_token_accuracy": 0.11605177596211433,
|
|
"num_tokens": 5706066.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.300340795516968,
|
|
"epoch": 0.33362598770851626,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997789953069354,
|
|
"loss": 7.1025,
|
|
"mean_token_accuracy": 0.11065095141530037,
|
|
"num_tokens": 5722100.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.327347612380981,
|
|
"epoch": 0.3346015022924593,
|
|
"grad_norm": 0.6796875,
|
|
"learning_rate": 0.0004997758677025989,
|
|
"loss": 7.2352,
|
|
"mean_token_accuracy": 0.10814947485923768,
|
|
"num_tokens": 5739068.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.340763235092163,
|
|
"epoch": 0.3355770168764023,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004997727181339658,
|
|
"loss": 7.2012,
|
|
"mean_token_accuracy": 0.10828282088041305,
|
|
"num_tokens": 5756147.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.36230993270874,
|
|
"epoch": 0.3365525314603453,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004997695466013438,
|
|
"loss": 7.1668,
|
|
"mean_token_accuracy": 0.1097688615322113,
|
|
"num_tokens": 5772157.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.358744096755982,
|
|
"epoch": 0.3375280460442884,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004997663531050427,
|
|
"loss": 7.2051,
|
|
"mean_token_accuracy": 0.11012494564056396,
|
|
"num_tokens": 5788128.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.300439834594727,
|
|
"epoch": 0.3385035606282314,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004997631376453749,
|
|
"loss": 7.1783,
|
|
"mean_token_accuracy": 0.11115790754556656,
|
|
"num_tokens": 5804793.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.296383476257324,
|
|
"epoch": 0.3394790752121744,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004997599002226541,
|
|
"loss": 7.1114,
|
|
"mean_token_accuracy": 0.11314733251929283,
|
|
"num_tokens": 5822409.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.250304651260376,
|
|
"epoch": 0.3404545897961175,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000499756640837197,
|
|
"loss": 7.1525,
|
|
"mean_token_accuracy": 0.1107781283557415,
|
|
"num_tokens": 5839023.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.371572017669678,
|
|
"epoch": 0.3414301043800605,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.000499753359489322,
|
|
"loss": 7.1678,
|
|
"mean_token_accuracy": 0.10706163644790649,
|
|
"num_tokens": 5855804.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.260924053192139,
|
|
"epoch": 0.3424056189640035,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004997500561793498,
|
|
"loss": 7.1683,
|
|
"mean_token_accuracy": 0.1108716905117035,
|
|
"num_tokens": 5872372.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.425952625274658,
|
|
"epoch": 0.3433811335479465,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004997467309076031,
|
|
"loss": 7.2114,
|
|
"mean_token_accuracy": 0.1058623306453228,
|
|
"num_tokens": 5888666.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.203307580947876,
|
|
"epoch": 0.3443566481318896,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0004997433836744069,
|
|
"loss": 7.1538,
|
|
"mean_token_accuracy": 0.1110640175640583,
|
|
"num_tokens": 5905577.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.353317165374756,
|
|
"epoch": 0.3453321627158326,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004997400144800882,
|
|
"loss": 7.1263,
|
|
"mean_token_accuracy": 0.11214198097586632,
|
|
"num_tokens": 5922570.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.180663681030273,
|
|
"epoch": 0.34630767729977563,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004997366233249762,
|
|
"loss": 7.0759,
|
|
"mean_token_accuracy": 0.1237415112555027,
|
|
"num_tokens": 5938995.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.281846189498902,
|
|
"epoch": 0.34728319188371864,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004997332102094024,
|
|
"loss": 7.095,
|
|
"mean_token_accuracy": 0.11867170929908752,
|
|
"num_tokens": 5954869.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.409283924102783,
|
|
"epoch": 0.3482587064676617,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004997297751337004,
|
|
"loss": 7.2056,
|
|
"mean_token_accuracy": 0.11266031563282013,
|
|
"num_tokens": 5972750.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.21663875579834,
|
|
"epoch": 0.34923422105160473,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.0004997263180982055,
|
|
"loss": 7.1203,
|
|
"mean_token_accuracy": 0.11525158882141114,
|
|
"num_tokens": 5989004.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.242398023605347,
|
|
"epoch": 0.35020973563554775,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.0004997228391032559,
|
|
"loss": 7.1336,
|
|
"mean_token_accuracy": 0.1106021262705326,
|
|
"num_tokens": 6004749.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.271965169906617,
|
|
"epoch": 0.35118525021949076,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004997193381491913,
|
|
"loss": 7.0186,
|
|
"mean_token_accuracy": 0.11608157455921173,
|
|
"num_tokens": 6022235.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.253506374359131,
|
|
"epoch": 0.35216076480343383,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000499715815236354,
|
|
"loss": 7.1938,
|
|
"mean_token_accuracy": 0.1065413273870945,
|
|
"num_tokens": 6039194.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.348612976074219,
|
|
"epoch": 0.35313627938737685,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.000499712270365088,
|
|
"loss": 7.0573,
|
|
"mean_token_accuracy": 0.11130550652742385,
|
|
"num_tokens": 6055376.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.285213661193848,
|
|
"epoch": 0.35411179397131987,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004997087035357398,
|
|
"loss": 7.1085,
|
|
"mean_token_accuracy": 0.11257182136178016,
|
|
"num_tokens": 6071105.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.185001993179322,
|
|
"epoch": 0.3550873085552629,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004997051147486582,
|
|
"loss": 7.0481,
|
|
"mean_token_accuracy": 0.11566854268312454,
|
|
"num_tokens": 6087964.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.231231546401977,
|
|
"epoch": 0.35606282313920595,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004997015040041935,
|
|
"loss": 7.1079,
|
|
"mean_token_accuracy": 0.10794939324259759,
|
|
"num_tokens": 6105245.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.218174982070923,
|
|
"epoch": 0.35703833772314897,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004996978713026986,
|
|
"loss": 7.0645,
|
|
"mean_token_accuracy": 0.11960886195302009,
|
|
"num_tokens": 6121736.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.306903123855591,
|
|
"epoch": 0.358013852307092,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996942166445287,
|
|
"loss": 7.1124,
|
|
"mean_token_accuracy": 0.11296163201332092,
|
|
"num_tokens": 6137751.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.246757125854492,
|
|
"epoch": 0.358989366891035,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004996905400300407,
|
|
"loss": 7.1722,
|
|
"mean_token_accuracy": 0.11314689144492149,
|
|
"num_tokens": 6154655.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.333488273620605,
|
|
"epoch": 0.35996488147497807,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000499686841459594,
|
|
"loss": 7.1761,
|
|
"mean_token_accuracy": 0.10916234776377678,
|
|
"num_tokens": 6172230.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.245674562454224,
|
|
"epoch": 0.3609403960589211,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004996831209335497,
|
|
"loss": 7.1167,
|
|
"mean_token_accuracy": 0.10682919770479202,
|
|
"num_tokens": 6188520.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.304710102081299,
|
|
"epoch": 0.3619159106428641,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004996793784522719,
|
|
"loss": 7.1444,
|
|
"mean_token_accuracy": 0.1107605941593647,
|
|
"num_tokens": 6205555.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.308576536178589,
|
|
"epoch": 0.3628914252268071,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004996756140161259,
|
|
"loss": 7.137,
|
|
"mean_token_accuracy": 0.11305367797613144,
|
|
"num_tokens": 6223548.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.237602376937867,
|
|
"epoch": 0.3638669398107502,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004996718276254797,
|
|
"loss": 7.1446,
|
|
"mean_token_accuracy": 0.10687385573983192,
|
|
"num_tokens": 6239266.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.311763906478882,
|
|
"epoch": 0.3648424543946932,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004996680192807032,
|
|
"loss": 7.1103,
|
|
"mean_token_accuracy": 0.11415550112724304,
|
|
"num_tokens": 6255020.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.2628981590271,
|
|
"epoch": 0.3658179689786362,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004996641889821686,
|
|
"loss": 7.0561,
|
|
"mean_token_accuracy": 0.10903556793928146,
|
|
"num_tokens": 6272774.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.241775321960449,
|
|
"epoch": 0.36679348356257924,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004996603367302502,
|
|
"loss": 7.0623,
|
|
"mean_token_accuracy": 0.11404155045747758,
|
|
"num_tokens": 6290478.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.243292140960693,
|
|
"epoch": 0.3677689981465223,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004996564625253244,
|
|
"loss": 7.1434,
|
|
"mean_token_accuracy": 0.11211170926690102,
|
|
"num_tokens": 6308003.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.27335000038147,
|
|
"epoch": 0.3687445127304653,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004996525663677697,
|
|
"loss": 7.1017,
|
|
"mean_token_accuracy": 0.10900750830769539,
|
|
"num_tokens": 6325748.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.2766975402832035,
|
|
"epoch": 0.36972002731440834,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000499648648257967,
|
|
"loss": 7.0383,
|
|
"mean_token_accuracy": 0.12426020875573159,
|
|
"num_tokens": 6344031.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.189545965194702,
|
|
"epoch": 0.37069554189835136,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000499644708196299,
|
|
"loss": 7.1028,
|
|
"mean_token_accuracy": 0.1133454717695713,
|
|
"num_tokens": 6359821.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.167187023162842,
|
|
"epoch": 0.3716710564822944,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004996407461831507,
|
|
"loss": 6.9456,
|
|
"mean_token_accuracy": 0.1260436162352562,
|
|
"num_tokens": 6376464.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.296348905563354,
|
|
"epoch": 0.37264657106623744,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004996367622189094,
|
|
"loss": 7.1198,
|
|
"mean_token_accuracy": 0.1087517760694027,
|
|
"num_tokens": 6392415.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.1976234912872314,
|
|
"epoch": 0.37362208565018046,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004996327563039643,
|
|
"loss": 7.1297,
|
|
"mean_token_accuracy": 0.10959314480423928,
|
|
"num_tokens": 6409151.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.263360643386841,
|
|
"epoch": 0.3745976002341235,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004996287284387068,
|
|
"loss": 7.0703,
|
|
"mean_token_accuracy": 0.11664813607931138,
|
|
"num_tokens": 6426997.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.25260739326477,
|
|
"epoch": 0.37557311481806654,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004996246786235306,
|
|
"loss": 7.0849,
|
|
"mean_token_accuracy": 0.11517348885536194,
|
|
"num_tokens": 6442871.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.1664557456970215,
|
|
"epoch": 0.37654862940200956,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004996206068588314,
|
|
"loss": 7.077,
|
|
"mean_token_accuracy": 0.11249651536345481,
|
|
"num_tokens": 6459677.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.309895324707031,
|
|
"epoch": 0.3775241439859526,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004996165131450071,
|
|
"loss": 7.2022,
|
|
"mean_token_accuracy": 0.10661727115511895,
|
|
"num_tokens": 6477239.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 7.2785297393798825,
|
|
"epoch": 0.37849965856989565,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004996123974824576,
|
|
"loss": 7.0671,
|
|
"mean_token_accuracy": 0.1197501078248024,
|
|
"num_tokens": 6493277.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.1475608348846436,
|
|
"epoch": 0.37947517315383866,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004996082598715852,
|
|
"loss": 7.0766,
|
|
"mean_token_accuracy": 0.11377091929316521,
|
|
"num_tokens": 6509204.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.211893129348755,
|
|
"epoch": 0.3804506877377817,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000499604100312794,
|
|
"loss": 6.9508,
|
|
"mean_token_accuracy": 0.12616619616746902,
|
|
"num_tokens": 6525486.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.141159534454346,
|
|
"epoch": 0.3814262023217247,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004995999188064908,
|
|
"loss": 7.1537,
|
|
"mean_token_accuracy": 0.10739106014370918,
|
|
"num_tokens": 6542227.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.345539331436157,
|
|
"epoch": 0.38240171690566777,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000499595715353084,
|
|
"loss": 7.2156,
|
|
"mean_token_accuracy": 0.1026027925312519,
|
|
"num_tokens": 6558951.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.271983861923218,
|
|
"epoch": 0.3833772314896108,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004995914899529844,
|
|
"loss": 7.1239,
|
|
"mean_token_accuracy": 0.11046183109283447,
|
|
"num_tokens": 6574696.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.303521347045899,
|
|
"epoch": 0.3843527460735538,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004995872426066048,
|
|
"loss": 7.1337,
|
|
"mean_token_accuracy": 0.1097158245742321,
|
|
"num_tokens": 6591628.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.247618675231934,
|
|
"epoch": 0.3853282606574968,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004995829733143602,
|
|
"loss": 7.0381,
|
|
"mean_token_accuracy": 0.11576716899871826,
|
|
"num_tokens": 6608107.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.224512672424316,
|
|
"epoch": 0.3863037752414399,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.000499578682076668,
|
|
"loss": 7.1625,
|
|
"mean_token_accuracy": 0.11688075736165046,
|
|
"num_tokens": 6626699.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.352358055114746,
|
|
"epoch": 0.3872792898253829,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004995743688939474,
|
|
"loss": 7.187,
|
|
"mean_token_accuracy": 0.09994767829775811,
|
|
"num_tokens": 6644540.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.222046613693237,
|
|
"epoch": 0.3882548044093259,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.0004995700337666199,
|
|
"loss": 7.1305,
|
|
"mean_token_accuracy": 0.10689185112714768,
|
|
"num_tokens": 6660653.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.299875974655151,
|
|
"epoch": 0.38923031899326893,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.000499565676695109,
|
|
"loss": 7.1156,
|
|
"mean_token_accuracy": 0.10829658955335617,
|
|
"num_tokens": 6677553.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.261008453369141,
|
|
"epoch": 0.390205833577212,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004995612976798406,
|
|
"loss": 7.1023,
|
|
"mean_token_accuracy": 0.10821453481912613,
|
|
"num_tokens": 6695274.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.187166833877564,
|
|
"epoch": 0.391181348161155,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004995568967212425,
|
|
"loss": 7.0511,
|
|
"mean_token_accuracy": 0.11570428013801574,
|
|
"num_tokens": 6711166.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.146482372283936,
|
|
"epoch": 0.39215686274509803,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004995524738197448,
|
|
"loss": 7.0174,
|
|
"mean_token_accuracy": 0.11494055837392807,
|
|
"num_tokens": 6728827.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 7.1518950939178465,
|
|
"epoch": 0.39313237732904105,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004995480289757796,
|
|
"loss": 6.9904,
|
|
"mean_token_accuracy": 0.12252861484885216,
|
|
"num_tokens": 6745901.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.176534938812256,
|
|
"epoch": 0.3941078919129841,
|
|
"grad_norm": 0.6640625,
|
|
"learning_rate": 0.0004995435621897815,
|
|
"loss": 6.974,
|
|
"mean_token_accuracy": 0.12089047506451607,
|
|
"num_tokens": 6762176.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 7.1775041103363035,
|
|
"epoch": 0.39508340649692714,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995390734621867,
|
|
"loss": 7.1308,
|
|
"mean_token_accuracy": 0.1080321915447712,
|
|
"num_tokens": 6779322.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 7.353971719741821,
|
|
"epoch": 0.39605892108087015,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004995345627934338,
|
|
"loss": 7.056,
|
|
"mean_token_accuracy": 0.11802642568945884,
|
|
"num_tokens": 6794293.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.9723522663116455,
|
|
"epoch": 0.39703443566481317,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004995300301839637,
|
|
"loss": 6.8772,
|
|
"mean_token_accuracy": 0.13384915441274642,
|
|
"num_tokens": 6812880.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 7.228035497665405,
|
|
"epoch": 0.39800995024875624,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004995254756342194,
|
|
"loss": 7.0977,
|
|
"mean_token_accuracy": 0.1094105526804924,
|
|
"num_tokens": 6829230.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 7.2402630805969235,
|
|
"epoch": 0.39898546483269925,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004995208991446458,
|
|
"loss": 7.0857,
|
|
"mean_token_accuracy": 0.11126495227217674,
|
|
"num_tokens": 6846787.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 7.175650978088379,
|
|
"epoch": 0.39996097941664227,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004995163007156902,
|
|
"loss": 6.9601,
|
|
"mean_token_accuracy": 0.1128228485584259,
|
|
"num_tokens": 6862990.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 7.260164356231689,
|
|
"epoch": 0.4009364940005853,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004995116803478018,
|
|
"loss": 7.1035,
|
|
"mean_token_accuracy": 0.11000586748123169,
|
|
"num_tokens": 6879170.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 7.151089239120483,
|
|
"epoch": 0.40191200858452836,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004995070380414321,
|
|
"loss": 7.089,
|
|
"mean_token_accuracy": 0.10915887728333473,
|
|
"num_tokens": 6895635.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.192745923995972,
|
|
"epoch": 0.4028875231684714,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000499502373797035,
|
|
"loss": 6.9965,
|
|
"mean_token_accuracy": 0.11772431656718255,
|
|
"num_tokens": 6912836.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 7.169298124313355,
|
|
"epoch": 0.4038630377524144,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000499497687615066,
|
|
"loss": 7.0363,
|
|
"mean_token_accuracy": 0.12014092653989791,
|
|
"num_tokens": 6929507.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 7.242495536804199,
|
|
"epoch": 0.4048385523363574,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000499492979495983,
|
|
"loss": 7.0704,
|
|
"mean_token_accuracy": 0.10755309760570526,
|
|
"num_tokens": 6946527.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 7.223427963256836,
|
|
"epoch": 0.4058140669203005,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004994882494402462,
|
|
"loss": 7.0539,
|
|
"mean_token_accuracy": 0.11421060264110565,
|
|
"num_tokens": 6962582.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 7.19104118347168,
|
|
"epoch": 0.4067895815042435,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004994834974483178,
|
|
"loss": 7.0277,
|
|
"mean_token_accuracy": 0.11575962379574775,
|
|
"num_tokens": 6979389.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 7.1313694477081295,
|
|
"epoch": 0.4077650960881865,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004994787235206621,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.11551673784852028,
|
|
"num_tokens": 6996637.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 7.241458940505981,
|
|
"epoch": 0.4087406106721295,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004994739276577456,
|
|
"loss": 7.1171,
|
|
"mean_token_accuracy": 0.10805290937423706,
|
|
"num_tokens": 7013013.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 7.170071363449097,
|
|
"epoch": 0.4097161252560726,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004994691098600369,
|
|
"loss": 7.0776,
|
|
"mean_token_accuracy": 0.11683185324072838,
|
|
"num_tokens": 7031476.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 7.2697829246521,
|
|
"epoch": 0.4106916398400156,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004994642701280068,
|
|
"loss": 6.9868,
|
|
"mean_token_accuracy": 0.11399012506008148,
|
|
"num_tokens": 7047424.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 7.187478828430176,
|
|
"epoch": 0.4116671544239586,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0004994594084621282,
|
|
"loss": 7.0799,
|
|
"mean_token_accuracy": 0.11093296855688095,
|
|
"num_tokens": 7063793.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 7.3022997856140135,
|
|
"epoch": 0.41264266900790164,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004994545248628762,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.10961232632398606,
|
|
"num_tokens": 7079026.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 7.164968585968017,
|
|
"epoch": 0.4136181835918447,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000499449619330728,
|
|
"loss": 7.0457,
|
|
"mean_token_accuracy": 0.11346296891570092,
|
|
"num_tokens": 7097524.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 7.20493860244751,
|
|
"epoch": 0.41459369817578773,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.000499444691866163,
|
|
"loss": 7.1198,
|
|
"mean_token_accuracy": 0.10945150628685951,
|
|
"num_tokens": 7115514.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 7.2198486328125,
|
|
"epoch": 0.41556921275973074,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004994397424696625,
|
|
"loss": 7.0593,
|
|
"mean_token_accuracy": 0.11467726454138756,
|
|
"num_tokens": 7132589.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 7.19329047203064,
|
|
"epoch": 0.4165447273436738,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004994347711417105,
|
|
"loss": 7.0692,
|
|
"mean_token_accuracy": 0.10977566614747047,
|
|
"num_tokens": 7149416.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 7.157116937637329,
|
|
"epoch": 0.41752024192761683,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004994297778827924,
|
|
"loss": 7.0036,
|
|
"mean_token_accuracy": 0.11760117933154106,
|
|
"num_tokens": 7165763.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 7.109394168853759,
|
|
"epoch": 0.41849575651155985,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004994247626933964,
|
|
"loss": 6.9874,
|
|
"mean_token_accuracy": 0.12286769747734069,
|
|
"num_tokens": 7183131.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 7.155841255187989,
|
|
"epoch": 0.41947127109550286,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004994197255740124,
|
|
"loss": 7.0219,
|
|
"mean_token_accuracy": 0.11699268147349358,
|
|
"num_tokens": 7200467.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 7.1444660186767575,
|
|
"epoch": 0.42044678567944593,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004994146665251327,
|
|
"loss": 7.0043,
|
|
"mean_token_accuracy": 0.11433120444417,
|
|
"num_tokens": 7217311.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 7.217905759811401,
|
|
"epoch": 0.42142230026338895,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004994095855472516,
|
|
"loss": 7.0209,
|
|
"mean_token_accuracy": 0.1175784558057785,
|
|
"num_tokens": 7235244.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 7.1584502220153805,
|
|
"epoch": 0.42239781484733196,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004994044826408656,
|
|
"loss": 7.0469,
|
|
"mean_token_accuracy": 0.11080644354224205,
|
|
"num_tokens": 7252235.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 7.209628391265869,
|
|
"epoch": 0.423373329431275,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004993993578064733,
|
|
"loss": 7.0503,
|
|
"mean_token_accuracy": 0.11199532449245453,
|
|
"num_tokens": 7269367.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 7.252705907821655,
|
|
"epoch": 0.42434884401521805,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004993942110445756,
|
|
"loss": 7.0388,
|
|
"mean_token_accuracy": 0.11196756064891815,
|
|
"num_tokens": 7287198.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 7.162608671188354,
|
|
"epoch": 0.42532435859916107,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004993890423556753,
|
|
"loss": 6.9824,
|
|
"mean_token_accuracy": 0.11178981512784958,
|
|
"num_tokens": 7302356.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 7.153040552139283,
|
|
"epoch": 0.4262998731831041,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004993838517402775,
|
|
"loss": 6.9713,
|
|
"mean_token_accuracy": 0.12076517716050147,
|
|
"num_tokens": 7318530.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 7.195984268188477,
|
|
"epoch": 0.4272753877670471,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.0004993786391988894,
|
|
"loss": 6.9641,
|
|
"mean_token_accuracy": 0.11528254747390747,
|
|
"num_tokens": 7335195.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 7.076996326446533,
|
|
"epoch": 0.42825090235099017,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004993734047320206,
|
|
"loss": 6.975,
|
|
"mean_token_accuracy": 0.11309642419219017,
|
|
"num_tokens": 7352233.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 7.198031425476074,
|
|
"epoch": 0.4292264169349332,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004993681483401821,
|
|
"loss": 7.0177,
|
|
"mean_token_accuracy": 0.1178817830979824,
|
|
"num_tokens": 7368569.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 7.138528823852539,
|
|
"epoch": 0.4302019315188762,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004993628700238879,
|
|
"loss": 7.0605,
|
|
"mean_token_accuracy": 0.11159829273819924,
|
|
"num_tokens": 7384669.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 7.243989372253418,
|
|
"epoch": 0.4311774461028192,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004993575697836537,
|
|
"loss": 7.0764,
|
|
"mean_token_accuracy": 0.10855081602931023,
|
|
"num_tokens": 7401833.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 7.177517652511597,
|
|
"epoch": 0.4321529606867623,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004993522476199975,
|
|
"loss": 6.9746,
|
|
"mean_token_accuracy": 0.11231087371706963,
|
|
"num_tokens": 7418663.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 7.1457499980926515,
|
|
"epoch": 0.4331284752707053,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004993469035334392,
|
|
"loss": 6.9562,
|
|
"mean_token_accuracy": 0.12267854809761047,
|
|
"num_tokens": 7434701.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 7.158255243301392,
|
|
"epoch": 0.4341039898546483,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000499341537524501,
|
|
"loss": 7.0339,
|
|
"mean_token_accuracy": 0.10957926958799362,
|
|
"num_tokens": 7450154.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 7.012423801422119,
|
|
"epoch": 0.43507950443859134,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004993361495937074,
|
|
"loss": 6.8259,
|
|
"mean_token_accuracy": 0.1353221409022808,
|
|
"num_tokens": 7468110.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 7.148668670654297,
|
|
"epoch": 0.4360550190225344,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004993307397415848,
|
|
"loss": 6.9687,
|
|
"mean_token_accuracy": 0.11352093443274498,
|
|
"num_tokens": 7482962.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.223632144927978,
|
|
"epoch": 0.4370305336064774,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.0004993253079686619,
|
|
"loss": 7.0961,
|
|
"mean_token_accuracy": 0.1094274289906025,
|
|
"num_tokens": 7499116.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 7.205734777450561,
|
|
"epoch": 0.43800604819042044,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004993198542754692,
|
|
"loss": 7.0599,
|
|
"mean_token_accuracy": 0.11345275118947029,
|
|
"num_tokens": 7515765.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 7.174308013916016,
|
|
"epoch": 0.43898156277436345,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004993143786625401,
|
|
"loss": 7.0388,
|
|
"mean_token_accuracy": 0.11509429663419724,
|
|
"num_tokens": 7532074.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 7.064720487594604,
|
|
"epoch": 0.4399570773583065,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004993088811304092,
|
|
"loss": 6.8781,
|
|
"mean_token_accuracy": 0.12545035257935525,
|
|
"num_tokens": 7547704.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 7.157081985473633,
|
|
"epoch": 0.44093259194224954,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004993033616796139,
|
|
"loss": 6.9942,
|
|
"mean_token_accuracy": 0.11518247872591018,
|
|
"num_tokens": 7563317.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 7.077218532562256,
|
|
"epoch": 0.44190810652619256,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.0004992978203106936,
|
|
"loss": 6.9864,
|
|
"mean_token_accuracy": 0.1106504775583744,
|
|
"num_tokens": 7580770.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 7.1526469707489015,
|
|
"epoch": 0.4428836211101356,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.0004992922570241896,
|
|
"loss": 6.9601,
|
|
"mean_token_accuracy": 0.11600121185183525,
|
|
"num_tokens": 7598181.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 7.17113733291626,
|
|
"epoch": 0.44385913569407864,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004992866718206457,
|
|
"loss": 6.9712,
|
|
"mean_token_accuracy": 0.11577035710215569,
|
|
"num_tokens": 7613739.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 7.134070205688476,
|
|
"epoch": 0.44483465027802166,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004992810647006077,
|
|
"loss": 7.0444,
|
|
"mean_token_accuracy": 0.1104633316397667,
|
|
"num_tokens": 7629297.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 7.236656618118286,
|
|
"epoch": 0.4458101648619647,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004992754356646234,
|
|
"loss": 7.0396,
|
|
"mean_token_accuracy": 0.11165096834301949,
|
|
"num_tokens": 7645273.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 7.185686111450195,
|
|
"epoch": 0.4467856794459077,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.0004992697847132427,
|
|
"loss": 6.986,
|
|
"mean_token_accuracy": 0.11109313443303108,
|
|
"num_tokens": 7661296.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.987871694564819,
|
|
"epoch": 0.44776119402985076,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.000499264111847018,
|
|
"loss": 6.9891,
|
|
"mean_token_accuracy": 0.11404722481966019,
|
|
"num_tokens": 7678658.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 7.233692026138305,
|
|
"epoch": 0.4487367086137938,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004992584170665037,
|
|
"loss": 7.0149,
|
|
"mean_token_accuracy": 0.11008156463503838,
|
|
"num_tokens": 7693678.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 7.112862920761108,
|
|
"epoch": 0.4497122231977368,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000499252700372256,
|
|
"loss": 7.0179,
|
|
"mean_token_accuracy": 0.11447776332497597,
|
|
"num_tokens": 7709526.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 7.164843320846558,
|
|
"epoch": 0.4506877377816798,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004992469617648337,
|
|
"loss": 6.9164,
|
|
"mean_token_accuracy": 0.11364173144102097,
|
|
"num_tokens": 7725055.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.968656778335571,
|
|
"epoch": 0.4516632523656229,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004992412012447976,
|
|
"loss": 6.8243,
|
|
"mean_token_accuracy": 0.1330489881336689,
|
|
"num_tokens": 7742047.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 7.127957963943482,
|
|
"epoch": 0.4526387669495659,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004992354188127105,
|
|
"loss": 6.9932,
|
|
"mean_token_accuracy": 0.11281251534819603,
|
|
"num_tokens": 7758283.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 7.13198881149292,
|
|
"epoch": 0.4536142815335089,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004992296144691375,
|
|
"loss": 6.9322,
|
|
"mean_token_accuracy": 0.11692915037274361,
|
|
"num_tokens": 7774660.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 7.0494366645812985,
|
|
"epoch": 0.454589796117452,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004992237882146456,
|
|
"loss": 6.9336,
|
|
"mean_token_accuracy": 0.12027623504400253,
|
|
"num_tokens": 7792147.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 7.085804891586304,
|
|
"epoch": 0.455565310701395,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004992179400498044,
|
|
"loss": 6.884,
|
|
"mean_token_accuracy": 0.11990112587809562,
|
|
"num_tokens": 7809021.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 7.126865386962891,
|
|
"epoch": 0.456540825285338,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004992120699751852,
|
|
"loss": 6.9547,
|
|
"mean_token_accuracy": 0.11495627984404563,
|
|
"num_tokens": 7825074.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.991373491287232,
|
|
"epoch": 0.45751633986928103,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004992061779913616,
|
|
"loss": 6.882,
|
|
"mean_token_accuracy": 0.12588541582226753,
|
|
"num_tokens": 7843413.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 7.136522054672241,
|
|
"epoch": 0.4584918544532241,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004992002640989094,
|
|
"loss": 6.9829,
|
|
"mean_token_accuracy": 0.10877103731036186,
|
|
"num_tokens": 7859874.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 7.0421689510345455,
|
|
"epoch": 0.4594673690371671,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004991943282984065,
|
|
"loss": 6.9006,
|
|
"mean_token_accuracy": 0.12029767408967018,
|
|
"num_tokens": 7877302.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 7.098082065582275,
|
|
"epoch": 0.46044288362111013,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004991883705904328,
|
|
"loss": 7.0204,
|
|
"mean_token_accuracy": 0.11330386027693748,
|
|
"num_tokens": 7893207.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 7.0267304420471195,
|
|
"epoch": 0.46141839820505315,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004991823909755705,
|
|
"loss": 6.8693,
|
|
"mean_token_accuracy": 0.12727298960089684,
|
|
"num_tokens": 7910239.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 7.135089111328125,
|
|
"epoch": 0.4623939127889962,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004991763894544041,
|
|
"loss": 6.9956,
|
|
"mean_token_accuracy": 0.11252414360642433,
|
|
"num_tokens": 7928030.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 7.087430810928344,
|
|
"epoch": 0.46336942737293924,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004991703660275199,
|
|
"loss": 6.9537,
|
|
"mean_token_accuracy": 0.11425257176160812,
|
|
"num_tokens": 7943653.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 7.1508338928222654,
|
|
"epoch": 0.46434494195688225,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004991643206955064,
|
|
"loss": 7.0187,
|
|
"mean_token_accuracy": 0.1106310598552227,
|
|
"num_tokens": 7960564.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 7.126855993270874,
|
|
"epoch": 0.46532045654082527,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004991582534589544,
|
|
"loss": 6.9782,
|
|
"mean_token_accuracy": 0.11728396564722061,
|
|
"num_tokens": 7978045.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 7.158732795715332,
|
|
"epoch": 0.46629597112476834,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004991521643184567,
|
|
"loss": 7.0161,
|
|
"mean_token_accuracy": 0.11397831588983535,
|
|
"num_tokens": 7995074.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 7.136742782592774,
|
|
"epoch": 0.46727148570871135,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004991460532746087,
|
|
"loss": 7.0158,
|
|
"mean_token_accuracy": 0.10939189791679382,
|
|
"num_tokens": 8011800.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 7.1472780227661135,
|
|
"epoch": 0.46824700029265437,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000499139920328007,
|
|
"loss": 7.001,
|
|
"mean_token_accuracy": 0.11125169917941094,
|
|
"num_tokens": 8028090.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 7.1227765560150145,
|
|
"epoch": 0.4692225148765974,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004991337654792512,
|
|
"loss": 6.9313,
|
|
"mean_token_accuracy": 0.11610379964113235,
|
|
"num_tokens": 8044140.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 7.104448366165161,
|
|
"epoch": 0.47019802946054046,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004991275887289427,
|
|
"loss": 6.8733,
|
|
"mean_token_accuracy": 0.11896544992923737,
|
|
"num_tokens": 8060030.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 7.1025842189788815,
|
|
"epoch": 0.47117354404448347,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004991213900776849,
|
|
"loss": 6.9444,
|
|
"mean_token_accuracy": 0.11614150553941727,
|
|
"num_tokens": 8075445.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 7.138435411453247,
|
|
"epoch": 0.4721490586284265,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004991151695260838,
|
|
"loss": 6.9406,
|
|
"mean_token_accuracy": 0.1131170354783535,
|
|
"num_tokens": 8093250.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 7.017556381225586,
|
|
"epoch": 0.4731245732123695,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000499108927074747,
|
|
"loss": 6.8782,
|
|
"mean_token_accuracy": 0.12963026240468026,
|
|
"num_tokens": 8109615.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 7.0512158393859865,
|
|
"epoch": 0.4741000877963126,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004991026627242846,
|
|
"loss": 6.9899,
|
|
"mean_token_accuracy": 0.11359846070408822,
|
|
"num_tokens": 8126591.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 7.101228904724121,
|
|
"epoch": 0.4750756023802556,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004990963764753088,
|
|
"loss": 6.9083,
|
|
"mean_token_accuracy": 0.11548377051949502,
|
|
"num_tokens": 8141360.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 7.109776735305786,
|
|
"epoch": 0.4760511169641986,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004990900683284337,
|
|
"loss": 7.002,
|
|
"mean_token_accuracy": 0.11828856319189071,
|
|
"num_tokens": 8159688.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 7.081469440460205,
|
|
"epoch": 0.4770266315481416,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004990837382842758,
|
|
"loss": 6.9589,
|
|
"mean_token_accuracy": 0.11527570113539695,
|
|
"num_tokens": 8176887.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.990096187591552,
|
|
"epoch": 0.4780021461320847,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004990773863434537,
|
|
"loss": 6.7791,
|
|
"mean_token_accuracy": 0.12774069532752036,
|
|
"num_tokens": 8193372.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 7.087004041671753,
|
|
"epoch": 0.4789776607160277,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004990710125065879,
|
|
"loss": 6.9977,
|
|
"mean_token_accuracy": 0.11214468628168106,
|
|
"num_tokens": 8209713.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 7.11400089263916,
|
|
"epoch": 0.4799531752999707,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004990646167743015,
|
|
"loss": 6.9882,
|
|
"mean_token_accuracy": 0.11344318985939025,
|
|
"num_tokens": 8226359.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 7.047367954254151,
|
|
"epoch": 0.48092868988391374,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004990581991472193,
|
|
"loss": 6.9662,
|
|
"mean_token_accuracy": 0.1149329237639904,
|
|
"num_tokens": 8243182.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 7.1153247356414795,
|
|
"epoch": 0.4819042044678568,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004990517596259684,
|
|
"loss": 6.9621,
|
|
"mean_token_accuracy": 0.11368496045470237,
|
|
"num_tokens": 8259185.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 7.141877031326294,
|
|
"epoch": 0.4828797190517998,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004990452982111781,
|
|
"loss": 7.0401,
|
|
"mean_token_accuracy": 0.11041902601718903,
|
|
"num_tokens": 8275941.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 7.108094882965088,
|
|
"epoch": 0.48385523363574284,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004990388149034797,
|
|
"loss": 6.9616,
|
|
"mean_token_accuracy": 0.11487062349915504,
|
|
"num_tokens": 8292535.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 7.117975378036499,
|
|
"epoch": 0.48483074821968586,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0004990323097035069,
|
|
"loss": 6.9325,
|
|
"mean_token_accuracy": 0.12068861201405526,
|
|
"num_tokens": 8310562.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 7.053023147583008,
|
|
"epoch": 0.48580626280362893,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004990257826118952,
|
|
"loss": 6.952,
|
|
"mean_token_accuracy": 0.10922210142016411,
|
|
"num_tokens": 8327241.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 7.062554359436035,
|
|
"epoch": 0.48678177738757195,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 0.0004990192336292824,
|
|
"loss": 6.8853,
|
|
"mean_token_accuracy": 0.12106603980064393,
|
|
"num_tokens": 8343516.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.936242532730103,
|
|
"epoch": 0.48775729197151496,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004990126627563087,
|
|
"loss": 6.7596,
|
|
"mean_token_accuracy": 0.1254474215209484,
|
|
"num_tokens": 8361809.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.997405242919922,
|
|
"epoch": 0.488732806555458,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004990060699936159,
|
|
"loss": 6.8241,
|
|
"mean_token_accuracy": 0.12973423004150392,
|
|
"num_tokens": 8378632.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 7.076214027404785,
|
|
"epoch": 0.48970832113940105,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004989994553418483,
|
|
"loss": 6.967,
|
|
"mean_token_accuracy": 0.10970980674028397,
|
|
"num_tokens": 8396201.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 7.017474126815796,
|
|
"epoch": 0.49068383572334406,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004989928188016522,
|
|
"loss": 6.857,
|
|
"mean_token_accuracy": 0.12434607595205308,
|
|
"num_tokens": 8412851.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.934906721115112,
|
|
"epoch": 0.4916593503072871,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004989861603736762,
|
|
"loss": 6.6964,
|
|
"mean_token_accuracy": 0.13276463598012925,
|
|
"num_tokens": 8430026.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 7.01360068321228,
|
|
"epoch": 0.49263486489123015,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004989794800585711,
|
|
"loss": 6.9332,
|
|
"mean_token_accuracy": 0.11373277679085732,
|
|
"num_tokens": 8446696.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 7.166025400161743,
|
|
"epoch": 0.49361037947517317,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004989727778569893,
|
|
"loss": 7.0107,
|
|
"mean_token_accuracy": 0.11518367677927017,
|
|
"num_tokens": 8464827.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 7.0952709197998045,
|
|
"epoch": 0.4945858940591162,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004989660537695859,
|
|
"loss": 6.9573,
|
|
"mean_token_accuracy": 0.11457037478685379,
|
|
"num_tokens": 8480185.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.976978826522827,
|
|
"epoch": 0.4955614086430592,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000498959307797018,
|
|
"loss": 6.8495,
|
|
"mean_token_accuracy": 0.12594428285956383,
|
|
"num_tokens": 8497511.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 7.057211256027221,
|
|
"epoch": 0.49653692322700227,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004989525399399447,
|
|
"loss": 6.9164,
|
|
"mean_token_accuracy": 0.11622816175222397,
|
|
"num_tokens": 8514394.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 7.087595653533936,
|
|
"epoch": 0.4975124378109453,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004989457501990275,
|
|
"loss": 6.9275,
|
|
"mean_token_accuracy": 0.11328752487897872,
|
|
"num_tokens": 8531659.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 7.078902673721314,
|
|
"epoch": 0.4984879523948883,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004989389385749297,
|
|
"loss": 6.968,
|
|
"mean_token_accuracy": 0.11704642698168755,
|
|
"num_tokens": 8547986.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 7.091986656188965,
|
|
"epoch": 0.4994634669788313,
|
|
"grad_norm": 0.6953125,
|
|
"learning_rate": 0.0004989321050683169,
|
|
"loss": 6.9336,
|
|
"mean_token_accuracy": 0.11741236075758935,
|
|
"num_tokens": 8565021.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 7.019861841201783,
|
|
"epoch": 0.5004389815627743,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000498925249679857,
|
|
"loss": 6.92,
|
|
"mean_token_accuracy": 0.11312920749187469,
|
|
"num_tokens": 8582135.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 7.050804996490479,
|
|
"epoch": 0.5014144961467174,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004989183724102196,
|
|
"loss": 6.9426,
|
|
"mean_token_accuracy": 0.11648973152041435,
|
|
"num_tokens": 8599946.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 7.095156240463257,
|
|
"epoch": 0.5023900107306605,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0004989114732600771,
|
|
"loss": 6.962,
|
|
"mean_token_accuracy": 0.11353963762521743,
|
|
"num_tokens": 8615831.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 7.0172281742095945,
|
|
"epoch": 0.5033655253146034,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004989045522301033,
|
|
"loss": 6.8659,
|
|
"mean_token_accuracy": 0.11678607761859894,
|
|
"num_tokens": 8631886.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 7.130041217803955,
|
|
"epoch": 0.5043410398985465,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004988976093209747,
|
|
"loss": 6.9024,
|
|
"mean_token_accuracy": 0.11544571220874786,
|
|
"num_tokens": 8647520.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 7.008956384658814,
|
|
"epoch": 0.5053165544824895,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004988906445333698,
|
|
"loss": 6.8257,
|
|
"mean_token_accuracy": 0.12740083709359168,
|
|
"num_tokens": 8663995.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 7.0500547885894775,
|
|
"epoch": 0.5062920690664325,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004988836578679691,
|
|
"loss": 6.8837,
|
|
"mean_token_accuracy": 0.11523761749267578,
|
|
"num_tokens": 8680201.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 7.057234573364258,
|
|
"epoch": 0.5072675836503756,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004988766493254552,
|
|
"loss": 6.8811,
|
|
"mean_token_accuracy": 0.12502829283475875,
|
|
"num_tokens": 8696978.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.975452184677124,
|
|
"epoch": 0.5082430982343186,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000498869618906513,
|
|
"loss": 6.9349,
|
|
"mean_token_accuracy": 0.1191725805401802,
|
|
"num_tokens": 8713335.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 7.085032987594604,
|
|
"epoch": 0.5092186128182616,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004988625666118296,
|
|
"loss": 6.9079,
|
|
"mean_token_accuracy": 0.12075944617390633,
|
|
"num_tokens": 8730722.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 7.009021377563476,
|
|
"epoch": 0.5101941274022047,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000498855492442094,
|
|
"loss": 6.8423,
|
|
"mean_token_accuracy": 0.11724851801991462,
|
|
"num_tokens": 8747430.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 7.055665969848633,
|
|
"epoch": 0.5111696419861477,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004988483963979975,
|
|
"loss": 6.9307,
|
|
"mean_token_accuracy": 0.1153571218252182,
|
|
"num_tokens": 8764817.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 7.095595455169677,
|
|
"epoch": 0.5121451565700907,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004988412784802336,
|
|
"loss": 6.952,
|
|
"mean_token_accuracy": 0.11426069363951682,
|
|
"num_tokens": 8780412.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 7.099550008773804,
|
|
"epoch": 0.5131206711540337,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004988341386894977,
|
|
"loss": 7.0133,
|
|
"mean_token_accuracy": 0.11473828405141831,
|
|
"num_tokens": 8795526.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 7.0228691577911375,
|
|
"epoch": 0.5140961857379768,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004988269770264876,
|
|
"loss": 6.8121,
|
|
"mean_token_accuracy": 0.12404470667243003,
|
|
"num_tokens": 8813568.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 7.096433353424072,
|
|
"epoch": 0.5150717003219198,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004988197934919029,
|
|
"loss": 6.975,
|
|
"mean_token_accuracy": 0.11357465535402297,
|
|
"num_tokens": 8831807.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 7.076529407501221,
|
|
"epoch": 0.5160472149058628,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004988125880864458,
|
|
"loss": 6.9645,
|
|
"mean_token_accuracy": 0.11619440764188767,
|
|
"num_tokens": 8848875.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.94493408203125,
|
|
"epoch": 0.5170227294898059,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 0.0004988053608108203,
|
|
"loss": 6.6698,
|
|
"mean_token_accuracy": 0.12810667157173156,
|
|
"num_tokens": 8866317.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 7.012036466598511,
|
|
"epoch": 0.517998244073749,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004987981116657325,
|
|
"loss": 6.8267,
|
|
"mean_token_accuracy": 0.11821717917919158,
|
|
"num_tokens": 8882543.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 7.038132619857788,
|
|
"epoch": 0.5189737586576919,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004987908406518909,
|
|
"loss": 6.8452,
|
|
"mean_token_accuracy": 0.11478555351495742,
|
|
"num_tokens": 8898640.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 7.044437074661255,
|
|
"epoch": 0.519949273241635,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.000498783547770006,
|
|
"loss": 6.9379,
|
|
"mean_token_accuracy": 0.11549510955810546,
|
|
"num_tokens": 8914649.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 7.070078182220459,
|
|
"epoch": 0.5209247878255779,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004987762330207903,
|
|
"loss": 6.9632,
|
|
"mean_token_accuracy": 0.11580834016203881,
|
|
"num_tokens": 8931585.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.993145561218261,
|
|
"epoch": 0.521900302409521,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004987688964049587,
|
|
"loss": 6.8512,
|
|
"mean_token_accuracy": 0.12186183482408523,
|
|
"num_tokens": 8948827.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.969137239456177,
|
|
"epoch": 0.5228758169934641,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.000498761537923228,
|
|
"loss": 6.8188,
|
|
"mean_token_accuracy": 0.12476989179849625,
|
|
"num_tokens": 8965680.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 7.103134489059448,
|
|
"epoch": 0.523851331577407,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.0004987541575763175,
|
|
"loss": 6.945,
|
|
"mean_token_accuracy": 0.10933230742812157,
|
|
"num_tokens": 8982813.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 7.053329753875732,
|
|
"epoch": 0.5248268461613501,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.000498746755364948,
|
|
"loss": 6.9238,
|
|
"mean_token_accuracy": 0.11738716661930085,
|
|
"num_tokens": 9000497.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 7.0302845478057865,
|
|
"epoch": 0.5258023607452932,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.0004987393312898431,
|
|
"loss": 6.852,
|
|
"mean_token_accuracy": 0.11841250211000443,
|
|
"num_tokens": 9016195.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.972050142288208,
|
|
"epoch": 0.5267778753292361,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004987318853517282,
|
|
"loss": 6.8136,
|
|
"mean_token_accuracy": 0.1275354102253914,
|
|
"num_tokens": 9033295.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 7.0185223579406735,
|
|
"epoch": 0.5277533899131792,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004987244175513308,
|
|
"loss": 6.9096,
|
|
"mean_token_accuracy": 0.11717701405286789,
|
|
"num_tokens": 9050725.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 7.055005502700806,
|
|
"epoch": 0.5287289044971223,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004987169278893807,
|
|
"loss": 6.8533,
|
|
"mean_token_accuracy": 0.12087640091776848,
|
|
"num_tokens": 9067842.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 7.043317222595215,
|
|
"epoch": 0.5297044190810652,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004987094163666098,
|
|
"loss": 6.8802,
|
|
"mean_token_accuracy": 0.11303822621703148,
|
|
"num_tokens": 9084602.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.984019565582275,
|
|
"epoch": 0.5306799336650083,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000498701882983752,
|
|
"loss": 6.8169,
|
|
"mean_token_accuracy": 0.12400035038590432,
|
|
"num_tokens": 9100530.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 7.082251310348511,
|
|
"epoch": 0.5316554482489513,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004986943277415435,
|
|
"loss": 6.9054,
|
|
"mean_token_accuracy": 0.12150466665625573,
|
|
"num_tokens": 9117304.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.934389734268189,
|
|
"epoch": 0.5326309628328944,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004986867506407225,
|
|
"loss": 6.8461,
|
|
"mean_token_accuracy": 0.13145331293344498,
|
|
"num_tokens": 9133974.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 7.023757410049439,
|
|
"epoch": 0.5336064774168374,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0004986791516820295,
|
|
"loss": 6.82,
|
|
"mean_token_accuracy": 0.12260211110115052,
|
|
"num_tokens": 9150600.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.958559894561768,
|
|
"epoch": 0.5345819920007804,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 0.000498671530866207,
|
|
"loss": 6.8195,
|
|
"mean_token_accuracy": 0.11772075891494752,
|
|
"num_tokens": 9167449.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.999980163574219,
|
|
"epoch": 0.5355575065847235,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004986638881939997,
|
|
"loss": 6.7508,
|
|
"mean_token_accuracy": 0.12078059241175651,
|
|
"num_tokens": 9184570.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 7.025358009338379,
|
|
"epoch": 0.5365330211686665,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004986562236661544,
|
|
"loss": 6.9478,
|
|
"mean_token_accuracy": 0.10966789573431016,
|
|
"num_tokens": 9200537.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 7.055627679824829,
|
|
"epoch": 0.5375085357526095,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.00049864853728342,
|
|
"loss": 6.8321,
|
|
"mean_token_accuracy": 0.12615425363183022,
|
|
"num_tokens": 9217252.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 7.011317777633667,
|
|
"epoch": 0.5384840503365526,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004986408290465476,
|
|
"loss": 6.8942,
|
|
"mean_token_accuracy": 0.11614497005939484,
|
|
"num_tokens": 9234108.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 7.009395360946655,
|
|
"epoch": 0.5394595649204955,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004986330989562906,
|
|
"loss": 6.79,
|
|
"mean_token_accuracy": 0.12128815576434135,
|
|
"num_tokens": 9251017.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.907582473754883,
|
|
"epoch": 0.5404350795044386,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004986253470134042,
|
|
"loss": 6.7643,
|
|
"mean_token_accuracy": 0.12897545844316483,
|
|
"num_tokens": 9267922.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.942857360839843,
|
|
"epoch": 0.5414105940883817,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.0004986175732186459,
|
|
"loss": 6.7594,
|
|
"mean_token_accuracy": 0.13300774469971657,
|
|
"num_tokens": 9284954.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 7.067489337921143,
|
|
"epoch": 0.5423861086723246,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 0.0004986097775727752,
|
|
"loss": 6.8754,
|
|
"mean_token_accuracy": 0.12139702215790749,
|
|
"num_tokens": 9303306.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.946069002151489,
|
|
"epoch": 0.5433616232562677,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0004986019600765541,
|
|
"loss": 6.7636,
|
|
"mean_token_accuracy": 0.12780117020010948,
|
|
"num_tokens": 9319754.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.997291612625122,
|
|
"epoch": 0.5443371378402108,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004985941207307462,
|
|
"loss": 6.8388,
|
|
"mean_token_accuracy": 0.12043605893850326,
|
|
"num_tokens": 9335736.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.960863447189331,
|
|
"epoch": 0.5453126524241537,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004985862595361179,
|
|
"loss": 6.8276,
|
|
"mean_token_accuracy": 0.1238748125731945,
|
|
"num_tokens": 9350435.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 7.038047218322754,
|
|
"epoch": 0.5462881670080968,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004985783764934372,
|
|
"loss": 6.8846,
|
|
"mean_token_accuracy": 0.11675451174378396,
|
|
"num_tokens": 9367613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.985388374328613,
|
|
"epoch": 0.5472636815920398,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004985704716034744,
|
|
"loss": 6.7739,
|
|
"mean_token_accuracy": 0.12371559590101242,
|
|
"num_tokens": 9385590.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 7.0262994289398195,
|
|
"epoch": 0.5482391961759828,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004985625448670019,
|
|
"loss": 6.9055,
|
|
"mean_token_accuracy": 0.10927185341715813,
|
|
"num_tokens": 9403286.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 7.065427780151367,
|
|
"epoch": 0.5492147107599259,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004985545962847943,
|
|
"loss": 6.8881,
|
|
"mean_token_accuracy": 0.11537767797708512,
|
|
"num_tokens": 9419880.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 7.033570194244385,
|
|
"epoch": 0.5501902253438689,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004985466258576283,
|
|
"loss": 6.8455,
|
|
"mean_token_accuracy": 0.11745240762829781,
|
|
"num_tokens": 9436579.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 7.046221017837524,
|
|
"epoch": 0.5511657399278119,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004985386335862829,
|
|
"loss": 6.93,
|
|
"mean_token_accuracy": 0.11016375422477723,
|
|
"num_tokens": 9452802.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 7.053445672988891,
|
|
"epoch": 0.552141254511755,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004985306194715387,
|
|
"loss": 6.938,
|
|
"mean_token_accuracy": 0.11385038048028946,
|
|
"num_tokens": 9468297.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 7.024582815170288,
|
|
"epoch": 0.553116769095698,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004985225835141792,
|
|
"loss": 6.9441,
|
|
"mean_token_accuracy": 0.11334621459245682,
|
|
"num_tokens": 9484954.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 7.109255123138428,
|
|
"epoch": 0.554092283679641,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004985145257149895,
|
|
"loss": 6.9819,
|
|
"mean_token_accuracy": 0.11194085255265236,
|
|
"num_tokens": 9502211.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 7.02985258102417,
|
|
"epoch": 0.555067798263584,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004985064460747569,
|
|
"loss": 6.8801,
|
|
"mean_token_accuracy": 0.11415814831852913,
|
|
"num_tokens": 9518103.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.930572509765625,
|
|
"epoch": 0.5560433128475271,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000498498344594271,
|
|
"loss": 6.8082,
|
|
"mean_token_accuracy": 0.1147056095302105,
|
|
"num_tokens": 9534155.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.953532028198242,
|
|
"epoch": 0.5570188274314701,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004984902212743234,
|
|
"loss": 6.8006,
|
|
"mean_token_accuracy": 0.12907804399728776,
|
|
"num_tokens": 9549972.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.932996416091919,
|
|
"epoch": 0.5579943420154131,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004984820761157079,
|
|
"loss": 6.7131,
|
|
"mean_token_accuracy": 0.1321914590895176,
|
|
"num_tokens": 9567306.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 7.030141401290893,
|
|
"epoch": 0.5589698565993562,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004984739091192205,
|
|
"loss": 6.9212,
|
|
"mean_token_accuracy": 0.1166323758661747,
|
|
"num_tokens": 9583488.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.894357776641845,
|
|
"epoch": 0.5599453711832992,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004984657202856592,
|
|
"loss": 6.698,
|
|
"mean_token_accuracy": 0.13180340230464935,
|
|
"num_tokens": 9600030.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.904035568237305,
|
|
"epoch": 0.5609208857672422,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000498457509615824,
|
|
"loss": 6.8267,
|
|
"mean_token_accuracy": 0.12177458480000496,
|
|
"num_tokens": 9615791.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 7.084134531021118,
|
|
"epoch": 0.5618964003511853,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004984492771105175,
|
|
"loss": 6.9083,
|
|
"mean_token_accuracy": 0.11326306611299515,
|
|
"num_tokens": 9630955.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.886215400695801,
|
|
"epoch": 0.5628719149351282,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004984410227705439,
|
|
"loss": 6.8119,
|
|
"mean_token_accuracy": 0.12833485156297683,
|
|
"num_tokens": 9647743.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.921821880340576,
|
|
"epoch": 0.5638474295190713,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0004984327465967101,
|
|
"loss": 6.7455,
|
|
"mean_token_accuracy": 0.12699681594967843,
|
|
"num_tokens": 9665495.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.982145643234253,
|
|
"epoch": 0.5648229441030144,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0004984244485898246,
|
|
"loss": 6.7774,
|
|
"mean_token_accuracy": 0.12316017523407936,
|
|
"num_tokens": 9683489.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.810558557510376,
|
|
"epoch": 0.5657984586869573,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.0004984161287506982,
|
|
"loss": 6.7105,
|
|
"mean_token_accuracy": 0.14188627004623414,
|
|
"num_tokens": 9700491.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 7.061842060089111,
|
|
"epoch": 0.5667739732709004,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000498407787080144,
|
|
"loss": 6.8763,
|
|
"mean_token_accuracy": 0.11491131633520127,
|
|
"num_tokens": 9715694.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 7.022202444076538,
|
|
"epoch": 0.5677494878548435,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000498399423578977,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.11375072970986366,
|
|
"num_tokens": 9731317.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 7.0759697437286375,
|
|
"epoch": 0.5687250024387864,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004983910382480146,
|
|
"loss": 6.9049,
|
|
"mean_token_accuracy": 0.1123342789709568,
|
|
"num_tokens": 9747928.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.998634338378906,
|
|
"epoch": 0.5697005170227295,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004983826310880762,
|
|
"loss": 6.841,
|
|
"mean_token_accuracy": 0.11529184207320213,
|
|
"num_tokens": 9765454.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 7.041535997390747,
|
|
"epoch": 0.5706760316066726,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004983742020999832,
|
|
"loss": 6.8941,
|
|
"mean_token_accuracy": 0.11383099108934402,
|
|
"num_tokens": 9782255.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.913339376449585,
|
|
"epoch": 0.5716515461906155,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.0004983657512845593,
|
|
"loss": 6.7069,
|
|
"mean_token_accuracy": 0.12910252287983895,
|
|
"num_tokens": 9798680.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.961237955093384,
|
|
"epoch": 0.5726270607745586,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004983572786426302,
|
|
"loss": 6.8695,
|
|
"mean_token_accuracy": 0.11734185367822647,
|
|
"num_tokens": 9813955.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.88823676109314,
|
|
"epoch": 0.5736025753585016,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004983487841750239,
|
|
"loss": 6.7594,
|
|
"mean_token_accuracy": 0.12689361944794655,
|
|
"num_tokens": 9830692.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 7.012313938140869,
|
|
"epoch": 0.5745780899424446,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004983402678825705,
|
|
"loss": 6.8604,
|
|
"mean_token_accuracy": 0.11689006462693215,
|
|
"num_tokens": 9848825.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 7.009697818756104,
|
|
"epoch": 0.5755536045263877,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004983317297661021,
|
|
"loss": 6.8636,
|
|
"mean_token_accuracy": 0.11862565949559212,
|
|
"num_tokens": 9865802.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.989213132858277,
|
|
"epoch": 0.5765291191103307,
|
|
"grad_norm": 0.72265625,
|
|
"learning_rate": 0.0004983231698264531,
|
|
"loss": 6.9113,
|
|
"mean_token_accuracy": 0.11293623223900795,
|
|
"num_tokens": 9883540.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 7.035774040222168,
|
|
"epoch": 0.5775046336942737,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004983145880644597,
|
|
"loss": 6.845,
|
|
"mean_token_accuracy": 0.11713336035609245,
|
|
"num_tokens": 9897724.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.980173540115357,
|
|
"epoch": 0.5784801482782168,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.0004983059844809608,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.12332117110490799,
|
|
"num_tokens": 9914710.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.9785926818847654,
|
|
"epoch": 0.5794556628621598,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0004982973590767969,
|
|
"loss": 6.905,
|
|
"mean_token_accuracy": 0.1224606953561306,
|
|
"num_tokens": 9932362.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.90735092163086,
|
|
"epoch": 0.5804311774461028,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000498288711852811,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.12667495682835578,
|
|
"num_tokens": 9949409.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.98042368888855,
|
|
"epoch": 0.5814066920300458,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000498280042809848,
|
|
"loss": 6.8482,
|
|
"mean_token_accuracy": 0.11787872463464737,
|
|
"num_tokens": 9967467.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.998588705062867,
|
|
"epoch": 0.5823822066139889,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004982713519487549,
|
|
"loss": 6.7842,
|
|
"mean_token_accuracy": 0.12298471853137016,
|
|
"num_tokens": 9983082.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.938471794128418,
|
|
"epoch": 0.5833577211979319,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0004982626392703812,
|
|
"loss": 6.858,
|
|
"mean_token_accuracy": 0.11961809620261192,
|
|
"num_tokens": 10000046.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.994626903533936,
|
|
"epoch": 0.5843332357818749,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000498253904775578,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.11904862523078918,
|
|
"num_tokens": 10014674.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.878867435455322,
|
|
"epoch": 0.585308750365818,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004982451484651988,
|
|
"loss": 6.6971,
|
|
"mean_token_accuracy": 0.1258040711283684,
|
|
"num_tokens": 10030647.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.585308750365818,
|
|
"eval_entropy": 6.76918303995275,
|
|
"eval_loss": 6.83972692489624,
|
|
"eval_mean_token_accuracy": 0.12215694767866397,
|
|
"eval_num_tokens": 10030647.0,
|
|
"eval_runtime": 11.9815,
|
|
"eval_samples_per_second": 1383.553,
|
|
"eval_steps_per_second": 173.017,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 51250,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2013648974807040.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|