4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.036678553234845,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691542720794677,
|
|
"epoch": 0.005094243504839531,
|
|
"grad_norm": 13.0625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8166,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12235.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691610050201415,
|
|
"epoch": 0.010188487009679063,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7442,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 24341.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691400146484375,
|
|
"epoch": 0.015282730514518594,
|
|
"grad_norm": 9.3125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4753,
|
|
"mean_token_accuracy": 0.015511087898630649,
|
|
"num_tokens": 34505.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.685736179351807,
|
|
"epoch": 0.020376974019358125,
|
|
"grad_norm": 5.71875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1943,
|
|
"mean_token_accuracy": 0.040183568745851515,
|
|
"num_tokens": 45816.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.646889209747314,
|
|
"epoch": 0.02547121752419766,
|
|
"grad_norm": 4.1875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9184,
|
|
"mean_token_accuracy": 0.043247510120272635,
|
|
"num_tokens": 57292.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.587969875335693,
|
|
"epoch": 0.030565461029037188,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7684,
|
|
"mean_token_accuracy": 0.042815128713846205,
|
|
"num_tokens": 70047.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.605268955230713,
|
|
"epoch": 0.03565970453387672,
|
|
"grad_norm": 2.953125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6701,
|
|
"mean_token_accuracy": 0.040725665539503096,
|
|
"num_tokens": 82218.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.603697872161865,
|
|
"epoch": 0.04075394803871625,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6447,
|
|
"mean_token_accuracy": 0.04414439275860786,
|
|
"num_tokens": 94126.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.564508724212647,
|
|
"epoch": 0.04584819154355578,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6114,
|
|
"mean_token_accuracy": 0.04162414483726025,
|
|
"num_tokens": 106527.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.579156398773193,
|
|
"epoch": 0.05094243504839532,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5471,
|
|
"mean_token_accuracy": 0.042017699033021924,
|
|
"num_tokens": 118284.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.582360935211181,
|
|
"epoch": 0.056036678553234846,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5311,
|
|
"mean_token_accuracy": 0.04019187185913324,
|
|
"num_tokens": 129462.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.567283725738525,
|
|
"epoch": 0.061130922058074376,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4374,
|
|
"mean_token_accuracy": 0.05231569707393646,
|
|
"num_tokens": 140488.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.469735717773437,
|
|
"epoch": 0.06622516556291391,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3729,
|
|
"mean_token_accuracy": 0.05711814947426319,
|
|
"num_tokens": 151931.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.409452629089355,
|
|
"epoch": 0.07131940906775344,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2764,
|
|
"mean_token_accuracy": 0.05579246319830418,
|
|
"num_tokens": 162824.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.449247074127197,
|
|
"epoch": 0.07641365257259297,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1642,
|
|
"mean_token_accuracy": 0.0578952107578516,
|
|
"num_tokens": 173502.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.389222621917725,
|
|
"epoch": 0.0815078960774325,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0716,
|
|
"mean_token_accuracy": 0.06245813779532909,
|
|
"num_tokens": 183899.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.330663871765136,
|
|
"epoch": 0.08660213958227203,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9753,
|
|
"mean_token_accuracy": 0.0661496214568615,
|
|
"num_tokens": 195359.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.337136840820312,
|
|
"epoch": 0.09169638308711156,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9478,
|
|
"mean_token_accuracy": 0.06206730008125305,
|
|
"num_tokens": 206472.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.36112995147705,
|
|
"epoch": 0.09679062659195109,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8421,
|
|
"mean_token_accuracy": 0.0680637639015913,
|
|
"num_tokens": 217846.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.245120906829834,
|
|
"epoch": 0.10188487009679063,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7311,
|
|
"mean_token_accuracy": 0.06576892882585525,
|
|
"num_tokens": 229036.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.20720853805542,
|
|
"epoch": 0.10697911360163016,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7046,
|
|
"mean_token_accuracy": 0.06441152691841126,
|
|
"num_tokens": 239838.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.15765905380249,
|
|
"epoch": 0.11207335710646969,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5953,
|
|
"mean_token_accuracy": 0.06702850237488747,
|
|
"num_tokens": 252006.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.032728290557861,
|
|
"epoch": 0.11716760061130922,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4248,
|
|
"mean_token_accuracy": 0.07769310809671878,
|
|
"num_tokens": 264027.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.969039916992188,
|
|
"epoch": 0.12226184411614875,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.325,
|
|
"mean_token_accuracy": 0.06973518244922161,
|
|
"num_tokens": 275156.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.869379806518555,
|
|
"epoch": 0.1273560876209883,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2347,
|
|
"mean_token_accuracy": 0.07396927624940872,
|
|
"num_tokens": 286240.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.799054813385009,
|
|
"epoch": 0.13245033112582782,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.1635,
|
|
"mean_token_accuracy": 0.07237125597894192,
|
|
"num_tokens": 298136.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.642311954498291,
|
|
"epoch": 0.13754457463066735,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.0224,
|
|
"mean_token_accuracy": 0.07531313709914685,
|
|
"num_tokens": 309682.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.317307949066162,
|
|
"epoch": 0.14263881813550688,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.884,
|
|
"mean_token_accuracy": 0.08923598602414132,
|
|
"num_tokens": 321330.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.228383636474609,
|
|
"epoch": 0.1477330616403464,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7986,
|
|
"mean_token_accuracy": 0.08174594640731811,
|
|
"num_tokens": 333364.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.043547344207763,
|
|
"epoch": 0.15282730514518594,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.69,
|
|
"mean_token_accuracy": 0.087934460490942,
|
|
"num_tokens": 345994.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.743598556518554,
|
|
"epoch": 0.15792154865002547,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.6734,
|
|
"mean_token_accuracy": 0.08026086539030075,
|
|
"num_tokens": 357966.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.630519485473632,
|
|
"epoch": 0.163015792154865,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.6224,
|
|
"mean_token_accuracy": 0.08176329433917999,
|
|
"num_tokens": 368890.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.526641750335694,
|
|
"epoch": 0.16811003565970453,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5345,
|
|
"mean_token_accuracy": 0.08557244464755058,
|
|
"num_tokens": 379995.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.317513370513916,
|
|
"epoch": 0.17320427916454406,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5155,
|
|
"mean_token_accuracy": 0.08169724568724632,
|
|
"num_tokens": 392482.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.162440204620362,
|
|
"epoch": 0.1782985226693836,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5077,
|
|
"mean_token_accuracy": 0.08479384928941727,
|
|
"num_tokens": 404318.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.083959865570069,
|
|
"epoch": 0.18339276617422312,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4791,
|
|
"mean_token_accuracy": 0.08923040181398392,
|
|
"num_tokens": 416223.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.061084938049316,
|
|
"epoch": 0.18848700967906265,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4398,
|
|
"mean_token_accuracy": 0.09084846526384353,
|
|
"num_tokens": 428375.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.927703142166138,
|
|
"epoch": 0.19358125318390218,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4848,
|
|
"mean_token_accuracy": 0.08455260321497918,
|
|
"num_tokens": 442032.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.927784204483032,
|
|
"epoch": 0.1986754966887417,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3906,
|
|
"mean_token_accuracy": 0.08499746695160866,
|
|
"num_tokens": 452898.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.933240032196045,
|
|
"epoch": 0.20376974019358127,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4295,
|
|
"mean_token_accuracy": 0.08968234360218048,
|
|
"num_tokens": 465037.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.818272733688355,
|
|
"epoch": 0.2088639836984208,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4109,
|
|
"mean_token_accuracy": 0.08541587293148041,
|
|
"num_tokens": 477816.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.907929992675781,
|
|
"epoch": 0.21395822720326033,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.4791,
|
|
"mean_token_accuracy": 0.08547281324863434,
|
|
"num_tokens": 489226.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.874213409423828,
|
|
"epoch": 0.21905247070809986,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4101,
|
|
"mean_token_accuracy": 0.08495720773935318,
|
|
"num_tokens": 501391.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.829102182388306,
|
|
"epoch": 0.22414671421293939,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4274,
|
|
"mean_token_accuracy": 0.08384364694356919,
|
|
"num_tokens": 513755.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.746364545822144,
|
|
"epoch": 0.22924095771777891,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3068,
|
|
"mean_token_accuracy": 0.09704143330454826,
|
|
"num_tokens": 525814.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.779020643234253,
|
|
"epoch": 0.23433520122261844,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3207,
|
|
"mean_token_accuracy": 0.09734901487827301,
|
|
"num_tokens": 537982.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.632878303527832,
|
|
"epoch": 0.23942944472745797,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2259,
|
|
"mean_token_accuracy": 0.09741544499993324,
|
|
"num_tokens": 549795.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.6897412776947025,
|
|
"epoch": 0.2445236882322975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2426,
|
|
"mean_token_accuracy": 0.09448553323745727,
|
|
"num_tokens": 561704.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.6036890029907225,
|
|
"epoch": 0.24961793173713703,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2715,
|
|
"mean_token_accuracy": 0.08808279559016227,
|
|
"num_tokens": 574736.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.690289926528931,
|
|
"epoch": 0.2547121752419766,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.357,
|
|
"mean_token_accuracy": 0.08886597231030464,
|
|
"num_tokens": 586785.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.658286476135254,
|
|
"epoch": 0.2598064187468161,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2521,
|
|
"mean_token_accuracy": 0.09355824813246727,
|
|
"num_tokens": 598738.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.630770540237426,
|
|
"epoch": 0.26490066225165565,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.273,
|
|
"mean_token_accuracy": 0.09633586630225181,
|
|
"num_tokens": 611431.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.61913652420044,
|
|
"epoch": 0.26999490575649515,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2633,
|
|
"mean_token_accuracy": 0.08988647237420082,
|
|
"num_tokens": 623253.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.583880138397217,
|
|
"epoch": 0.2750891492613347,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2567,
|
|
"mean_token_accuracy": 0.09057076498866082,
|
|
"num_tokens": 634663.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.594904994964599,
|
|
"epoch": 0.2801833927661742,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1883,
|
|
"mean_token_accuracy": 0.1007538914680481,
|
|
"num_tokens": 644578.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.535658311843872,
|
|
"epoch": 0.28527763627101377,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2628,
|
|
"mean_token_accuracy": 0.08800739347934723,
|
|
"num_tokens": 656226.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.578098011016846,
|
|
"epoch": 0.29037187977585327,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1927,
|
|
"mean_token_accuracy": 0.0906010314822197,
|
|
"num_tokens": 668863.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.590853548049926,
|
|
"epoch": 0.2954661232806928,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1582,
|
|
"mean_token_accuracy": 0.09744800627231598,
|
|
"num_tokens": 679993.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.549949026107788,
|
|
"epoch": 0.30056036678553233,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2245,
|
|
"mean_token_accuracy": 0.08970160409808159,
|
|
"num_tokens": 692197.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.570124864578247,
|
|
"epoch": 0.3056546102903719,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2664,
|
|
"mean_token_accuracy": 0.08972233831882477,
|
|
"num_tokens": 703363.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.577747011184693,
|
|
"epoch": 0.3107488537952114,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2417,
|
|
"mean_token_accuracy": 0.09180614426732063,
|
|
"num_tokens": 714335.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.526177406311035,
|
|
"epoch": 0.31584309730005095,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2336,
|
|
"mean_token_accuracy": 0.08975826352834701,
|
|
"num_tokens": 726638.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.562946748733521,
|
|
"epoch": 0.32093734080489045,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1228,
|
|
"mean_token_accuracy": 0.09438190683722496,
|
|
"num_tokens": 737115.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.465821599960327,
|
|
"epoch": 0.32603158430973,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1335,
|
|
"mean_token_accuracy": 0.09490179568529129,
|
|
"num_tokens": 748999.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.5147120475769045,
|
|
"epoch": 0.33112582781456956,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1197,
|
|
"mean_token_accuracy": 0.09471622779965401,
|
|
"num_tokens": 759858.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.490925359725952,
|
|
"epoch": 0.33622007131940906,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0829,
|
|
"mean_token_accuracy": 0.09656840115785599,
|
|
"num_tokens": 770636.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.387595987319946,
|
|
"epoch": 0.3413143148242486,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1451,
|
|
"mean_token_accuracy": 0.09274626299738883,
|
|
"num_tokens": 783221.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.407729530334473,
|
|
"epoch": 0.3464085583290881,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0896,
|
|
"mean_token_accuracy": 0.09786387607455253,
|
|
"num_tokens": 793548.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.483667325973511,
|
|
"epoch": 0.3515028018339277,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.093,
|
|
"mean_token_accuracy": 0.09398990496993065,
|
|
"num_tokens": 805553.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.409384870529175,
|
|
"epoch": 0.3565970453387672,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1612,
|
|
"mean_token_accuracy": 0.08771809116005898,
|
|
"num_tokens": 818475.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.439428186416626,
|
|
"epoch": 0.36169128884360674,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0588,
|
|
"mean_token_accuracy": 0.1010989487171173,
|
|
"num_tokens": 830826.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.413034200668335,
|
|
"epoch": 0.36678553234844624,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.126,
|
|
"mean_token_accuracy": 0.09531685337424278,
|
|
"num_tokens": 842850.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.3999276638031,
|
|
"epoch": 0.3718797758532858,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0959,
|
|
"mean_token_accuracy": 0.08785031288862229,
|
|
"num_tokens": 855954.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.448844003677368,
|
|
"epoch": 0.3769740193581253,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.08820546790957451,
|
|
"num_tokens": 867986.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.3457808017730715,
|
|
"epoch": 0.38206826286296486,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0625,
|
|
"mean_token_accuracy": 0.10003451183438301,
|
|
"num_tokens": 880063.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.308112478256225,
|
|
"epoch": 0.38716250636780436,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.069,
|
|
"mean_token_accuracy": 0.0941434882581234,
|
|
"num_tokens": 893320.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.378047084808349,
|
|
"epoch": 0.3922567498726439,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9678,
|
|
"mean_token_accuracy": 0.10630032047629356,
|
|
"num_tokens": 905118.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.412222957611084,
|
|
"epoch": 0.3973509933774834,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0692,
|
|
"mean_token_accuracy": 0.09440404176712036,
|
|
"num_tokens": 916403.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.367623662948608,
|
|
"epoch": 0.402445236882323,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1131,
|
|
"mean_token_accuracy": 0.09435304701328277,
|
|
"num_tokens": 927811.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.2225226879119875,
|
|
"epoch": 0.40753948038716253,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9568,
|
|
"mean_token_accuracy": 0.10003332197666168,
|
|
"num_tokens": 940095.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.278196573257446,
|
|
"epoch": 0.41263372389200204,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.10148229897022247,
|
|
"num_tokens": 951985.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.2614727973937985,
|
|
"epoch": 0.4177279673968416,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0128,
|
|
"mean_token_accuracy": 0.09977934956550598,
|
|
"num_tokens": 963981.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.3839428424835205,
|
|
"epoch": 0.4228222109016811,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1004,
|
|
"mean_token_accuracy": 0.09160929396748543,
|
|
"num_tokens": 976389.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.311860656738281,
|
|
"epoch": 0.42791645440652065,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9943,
|
|
"mean_token_accuracy": 0.09845278114080429,
|
|
"num_tokens": 987712.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.24138240814209,
|
|
"epoch": 0.43301069791136015,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9873,
|
|
"mean_token_accuracy": 0.09760711789131164,
|
|
"num_tokens": 998407.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.359812211990357,
|
|
"epoch": 0.4381049414161997,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0515,
|
|
"mean_token_accuracy": 0.09389390870928764,
|
|
"num_tokens": 1010275.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.249120759963989,
|
|
"epoch": 0.4431991849210392,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9664,
|
|
"mean_token_accuracy": 0.09880583584308625,
|
|
"num_tokens": 1022357.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.109994220733642,
|
|
"epoch": 0.44829342842587877,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8904,
|
|
"mean_token_accuracy": 0.10976919382810593,
|
|
"num_tokens": 1033751.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.278148746490478,
|
|
"epoch": 0.4533876719307183,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9559,
|
|
"mean_token_accuracy": 0.1025618351995945,
|
|
"num_tokens": 1045464.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.157083129882812,
|
|
"epoch": 0.45848191543555783,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9172,
|
|
"mean_token_accuracy": 0.10281323045492172,
|
|
"num_tokens": 1057281.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.2530436515808105,
|
|
"epoch": 0.46357615894039733,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.10712229833006859,
|
|
"num_tokens": 1070169.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.085860013961792,
|
|
"epoch": 0.4686704024452369,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.11112349703907967,
|
|
"num_tokens": 1081207.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.180644798278808,
|
|
"epoch": 0.4737646459500764,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9797,
|
|
"mean_token_accuracy": 0.0991374671459198,
|
|
"num_tokens": 1093846.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.150368309020996,
|
|
"epoch": 0.47885888945491595,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9856,
|
|
"mean_token_accuracy": 0.10191999301314354,
|
|
"num_tokens": 1106360.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.162285184860229,
|
|
"epoch": 0.48395313295975545,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8428,
|
|
"mean_token_accuracy": 0.11239263117313385,
|
|
"num_tokens": 1118042.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.218068408966064,
|
|
"epoch": 0.489047376464595,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8701,
|
|
"mean_token_accuracy": 0.10295976549386979,
|
|
"num_tokens": 1129625.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.12045316696167,
|
|
"epoch": 0.49414161996943456,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8821,
|
|
"mean_token_accuracy": 0.10984387025237083,
|
|
"num_tokens": 1140416.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.149784946441651,
|
|
"epoch": 0.49923586347427407,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8669,
|
|
"mean_token_accuracy": 0.10825323984026909,
|
|
"num_tokens": 1152273.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.108687162399292,
|
|
"epoch": 0.5043301069791136,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.823,
|
|
"mean_token_accuracy": 0.10351424664258957,
|
|
"num_tokens": 1163621.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.144402980804443,
|
|
"epoch": 0.5094243504839532,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9754,
|
|
"mean_token_accuracy": 0.0984008178114891,
|
|
"num_tokens": 1174445.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.111788082122803,
|
|
"epoch": 0.5145185939887926,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8904,
|
|
"mean_token_accuracy": 0.1037333846092224,
|
|
"num_tokens": 1187816.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.0931559085845945,
|
|
"epoch": 0.5196128374936322,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.10965610668063164,
|
|
"num_tokens": 1200523.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.0524476051330565,
|
|
"epoch": 0.5247070809984717,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8356,
|
|
"mean_token_accuracy": 0.10787132382392883,
|
|
"num_tokens": 1211841.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.0291201114654545,
|
|
"epoch": 0.5298013245033113,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7972,
|
|
"mean_token_accuracy": 0.11316964104771614,
|
|
"num_tokens": 1224076.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.041120100021362,
|
|
"epoch": 0.5348955680081507,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8326,
|
|
"mean_token_accuracy": 0.10762132853269576,
|
|
"num_tokens": 1236079.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.104992914199829,
|
|
"epoch": 0.5399898115129903,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8411,
|
|
"mean_token_accuracy": 0.10477398708462715,
|
|
"num_tokens": 1248922.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.033631038665772,
|
|
"epoch": 0.5450840550178299,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8513,
|
|
"mean_token_accuracy": 0.10244830325245857,
|
|
"num_tokens": 1260471.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.072809505462646,
|
|
"epoch": 0.5501782985226694,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9035,
|
|
"mean_token_accuracy": 0.1032500334084034,
|
|
"num_tokens": 1272295.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.1425238132476805,
|
|
"epoch": 0.5552725420275089,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8666,
|
|
"mean_token_accuracy": 0.10251828804612159,
|
|
"num_tokens": 1285056.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.086905145645142,
|
|
"epoch": 0.5603667855323484,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8308,
|
|
"mean_token_accuracy": 0.1039289228618145,
|
|
"num_tokens": 1296371.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.003297996520996,
|
|
"epoch": 0.565461029037188,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.807,
|
|
"mean_token_accuracy": 0.11644907146692277,
|
|
"num_tokens": 1308428.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.105059385299683,
|
|
"epoch": 0.5705552725420275,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8967,
|
|
"mean_token_accuracy": 0.10284779965877533,
|
|
"num_tokens": 1320586.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.042512512207031,
|
|
"epoch": 0.5756495160468671,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8834,
|
|
"mean_token_accuracy": 0.0991224467754364,
|
|
"num_tokens": 1332162.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.120642185211182,
|
|
"epoch": 0.5807437595517065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.8552,
|
|
"mean_token_accuracy": 0.11167115345597267,
|
|
"num_tokens": 1343574.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.9979644298553465,
|
|
"epoch": 0.5858380030565461,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7303,
|
|
"mean_token_accuracy": 0.10642987117171288,
|
|
"num_tokens": 1353970.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.101837110519409,
|
|
"epoch": 0.5909322465613857,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8664,
|
|
"mean_token_accuracy": 0.10311477482318879,
|
|
"num_tokens": 1366324.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.087606000900268,
|
|
"epoch": 0.5960264900662252,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8881,
|
|
"mean_token_accuracy": 0.10560108572244645,
|
|
"num_tokens": 1378778.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.984965896606445,
|
|
"epoch": 0.6011207335710647,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8006,
|
|
"mean_token_accuracy": 0.10560821816325187,
|
|
"num_tokens": 1390305.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.993460321426392,
|
|
"epoch": 0.6062149770759042,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7907,
|
|
"mean_token_accuracy": 0.11571490839123726,
|
|
"num_tokens": 1401122.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.989016771316528,
|
|
"epoch": 0.6113092205807438,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7646,
|
|
"mean_token_accuracy": 0.10929581746459008,
|
|
"num_tokens": 1413598.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.932051324844361,
|
|
"epoch": 0.6164034640855833,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.7509,
|
|
"mean_token_accuracy": 0.11002198383212089,
|
|
"num_tokens": 1425344.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.981126356124878,
|
|
"epoch": 0.6214977075904228,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.777,
|
|
"mean_token_accuracy": 0.1079341247677803,
|
|
"num_tokens": 1439096.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.983417558670044,
|
|
"epoch": 0.6265919510952623,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.888,
|
|
"mean_token_accuracy": 0.1041940689086914,
|
|
"num_tokens": 1451207.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.081508731842041,
|
|
"epoch": 0.6316861946001019,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8133,
|
|
"mean_token_accuracy": 0.10688796490430832,
|
|
"num_tokens": 1463404.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.853088903427124,
|
|
"epoch": 0.6367804381049414,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6653,
|
|
"mean_token_accuracy": 0.11994123682379723,
|
|
"num_tokens": 1474609.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.021486139297485,
|
|
"epoch": 0.6418746816097809,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.8337,
|
|
"mean_token_accuracy": 0.10318783521652222,
|
|
"num_tokens": 1486323.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.962870025634766,
|
|
"epoch": 0.6469689251146205,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7065,
|
|
"mean_token_accuracy": 0.11299853473901748,
|
|
"num_tokens": 1497325.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.9018639087677,
|
|
"epoch": 0.65206316861946,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6859,
|
|
"mean_token_accuracy": 0.11526524797081947,
|
|
"num_tokens": 1507329.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.870681667327881,
|
|
"epoch": 0.6571574121242996,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7217,
|
|
"mean_token_accuracy": 0.1096078835427761,
|
|
"num_tokens": 1518927.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.879950332641601,
|
|
"epoch": 0.6622516556291391,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7646,
|
|
"mean_token_accuracy": 0.10798789262771606,
|
|
"num_tokens": 1529171.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.088048267364502,
|
|
"epoch": 0.6673458991339786,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.8146,
|
|
"mean_token_accuracy": 0.10387829095125198,
|
|
"num_tokens": 1541295.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.9415205955505375,
|
|
"epoch": 0.6724401426388181,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.8109,
|
|
"mean_token_accuracy": 0.1034116804599762,
|
|
"num_tokens": 1553865.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.960649061203003,
|
|
"epoch": 0.6775343861436577,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.6791,
|
|
"mean_token_accuracy": 0.1155511736869812,
|
|
"num_tokens": 1565092.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.91525239944458,
|
|
"epoch": 0.6826286296484972,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7498,
|
|
"mean_token_accuracy": 0.11686733365058899,
|
|
"num_tokens": 1576170.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.939774513244629,
|
|
"epoch": 0.6877228731533367,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.712,
|
|
"mean_token_accuracy": 0.11363177075982094,
|
|
"num_tokens": 1588121.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.846085166931152,
|
|
"epoch": 0.6928171166581762,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7328,
|
|
"mean_token_accuracy": 0.11368786618113517,
|
|
"num_tokens": 1601141.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.856701278686524,
|
|
"epoch": 0.6979113601630158,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6897,
|
|
"mean_token_accuracy": 0.11437584608793258,
|
|
"num_tokens": 1613188.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.929333019256592,
|
|
"epoch": 0.7030056036678554,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7176,
|
|
"mean_token_accuracy": 0.11833318173885346,
|
|
"num_tokens": 1625889.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.873142099380493,
|
|
"epoch": 0.7080998471726948,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6713,
|
|
"mean_token_accuracy": 0.1138796404004097,
|
|
"num_tokens": 1636350.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.760504055023193,
|
|
"epoch": 0.7131940906775344,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6654,
|
|
"mean_token_accuracy": 0.11773468181490898,
|
|
"num_tokens": 1648290.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.85955548286438,
|
|
"epoch": 0.7182883341823739,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5593,
|
|
"mean_token_accuracy": 0.11682811230421067,
|
|
"num_tokens": 1658781.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.8511137008667,
|
|
"epoch": 0.7233825776872135,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6976,
|
|
"mean_token_accuracy": 0.10959015414118767,
|
|
"num_tokens": 1671779.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.841222906112671,
|
|
"epoch": 0.7284768211920529,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6185,
|
|
"mean_token_accuracy": 0.11852296516299247,
|
|
"num_tokens": 1683821.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.794290447235108,
|
|
"epoch": 0.7335710646968925,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5934,
|
|
"mean_token_accuracy": 0.11842493042349815,
|
|
"num_tokens": 1695754.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.7905267715454105,
|
|
"epoch": 0.738665308201732,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7078,
|
|
"mean_token_accuracy": 0.10955497696995735,
|
|
"num_tokens": 1707398.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.793871116638184,
|
|
"epoch": 0.7437595517065716,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.62,
|
|
"mean_token_accuracy": 0.11535302624106407,
|
|
"num_tokens": 1719315.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.790361738204956,
|
|
"epoch": 0.7488537952114112,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6036,
|
|
"mean_token_accuracy": 0.1142238661646843,
|
|
"num_tokens": 1731406.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.791077375411987,
|
|
"epoch": 0.7539480387162506,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.11140788197517396,
|
|
"num_tokens": 1744941.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.831547403335572,
|
|
"epoch": 0.7590422822210902,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6789,
|
|
"mean_token_accuracy": 0.10744564235210419,
|
|
"num_tokens": 1757824.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.820816898345948,
|
|
"epoch": 0.7641365257259297,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6841,
|
|
"mean_token_accuracy": 0.1142085425555706,
|
|
"num_tokens": 1770050.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.666744232177734,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6528,
|
|
"mean_token_accuracy": 0.11462347880005837,
|
|
"num_tokens": 1782083.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.888738059997559,
|
|
"epoch": 0.7743250127356087,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6007,
|
|
"mean_token_accuracy": 0.11330613195896148,
|
|
"num_tokens": 1792812.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.704314184188843,
|
|
"epoch": 0.7794192562404483,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5961,
|
|
"mean_token_accuracy": 0.11205876171588898,
|
|
"num_tokens": 1804724.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.7446770668029785,
|
|
"epoch": 0.7845134997452878,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6132,
|
|
"mean_token_accuracy": 0.1153540588915348,
|
|
"num_tokens": 1817051.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.810240411758423,
|
|
"epoch": 0.7896077432501274,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5863,
|
|
"mean_token_accuracy": 0.11449561789631843,
|
|
"num_tokens": 1828328.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.6139240741729735,
|
|
"epoch": 0.7947019867549668,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5736,
|
|
"mean_token_accuracy": 0.12113288342952729,
|
|
"num_tokens": 1840117.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.745065593719483,
|
|
"epoch": 0.7997962302598064,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5358,
|
|
"mean_token_accuracy": 0.11861959546804428,
|
|
"num_tokens": 1851097.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.701609706878662,
|
|
"epoch": 0.804890473764646,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6026,
|
|
"mean_token_accuracy": 0.11956391111016273,
|
|
"num_tokens": 1862282.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.7710367202758786,
|
|
"epoch": 0.8099847172694855,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5871,
|
|
"mean_token_accuracy": 0.11796328648924828,
|
|
"num_tokens": 1874936.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.704710197448731,
|
|
"epoch": 0.8150789607743251,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6084,
|
|
"mean_token_accuracy": 0.11659725233912469,
|
|
"num_tokens": 1887429.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.666276550292968,
|
|
"epoch": 0.8201732042791645,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5884,
|
|
"mean_token_accuracy": 0.11542436331510544,
|
|
"num_tokens": 1899378.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.6405881404876705,
|
|
"epoch": 0.8252674477840041,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5236,
|
|
"mean_token_accuracy": 0.1210794910788536,
|
|
"num_tokens": 1910723.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.753037929534912,
|
|
"epoch": 0.8303616912888436,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5719,
|
|
"mean_token_accuracy": 0.1203953631222248,
|
|
"num_tokens": 1921913.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.668088006973266,
|
|
"epoch": 0.8354559347936832,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5662,
|
|
"mean_token_accuracy": 0.12285715714097023,
|
|
"num_tokens": 1934052.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.671568870544434,
|
|
"epoch": 0.8405501782985226,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6321,
|
|
"mean_token_accuracy": 0.11708329319953918,
|
|
"num_tokens": 1947512.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.6495912075042725,
|
|
"epoch": 0.8456444218033622,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5694,
|
|
"mean_token_accuracy": 0.11596633568406105,
|
|
"num_tokens": 1959569.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.701305866241455,
|
|
"epoch": 0.8507386653082017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5576,
|
|
"mean_token_accuracy": 0.12322757616639138,
|
|
"num_tokens": 1970159.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.678005695343018,
|
|
"epoch": 0.8558329088130413,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6221,
|
|
"mean_token_accuracy": 0.11528186425566674,
|
|
"num_tokens": 1982688.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.634650087356567,
|
|
"epoch": 0.8609271523178808,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5171,
|
|
"mean_token_accuracy": 0.11815884709358215,
|
|
"num_tokens": 1993733.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.780072736740112,
|
|
"epoch": 0.8660213958227203,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5662,
|
|
"mean_token_accuracy": 0.12087653055787087,
|
|
"num_tokens": 2004880.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.491039991378784,
|
|
"epoch": 0.8711156393275599,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5444,
|
|
"mean_token_accuracy": 0.11627500280737876,
|
|
"num_tokens": 2017018.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.73184871673584,
|
|
"epoch": 0.8762098828323994,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6137,
|
|
"mean_token_accuracy": 0.1149538166821003,
|
|
"num_tokens": 2028930.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.617103576660156,
|
|
"epoch": 0.8813041263372389,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.552,
|
|
"mean_token_accuracy": 0.1166415274143219,
|
|
"num_tokens": 2039846.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.699741840362549,
|
|
"epoch": 0.8863983698420784,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.5565,
|
|
"mean_token_accuracy": 0.1190009169280529,
|
|
"num_tokens": 2051021.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.739818954467774,
|
|
"epoch": 0.891492613346918,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6827,
|
|
"mean_token_accuracy": 0.10884234607219696,
|
|
"num_tokens": 2064862.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.554408264160156,
|
|
"epoch": 0.8965868568517575,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5121,
|
|
"mean_token_accuracy": 0.12005752176046372,
|
|
"num_tokens": 2076316.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.6821098804473875,
|
|
"epoch": 0.9016811003565971,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.552,
|
|
"mean_token_accuracy": 0.11650667190551758,
|
|
"num_tokens": 2088115.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.785771703720092,
|
|
"epoch": 0.9067753438614365,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.6924,
|
|
"mean_token_accuracy": 0.11133024618029594,
|
|
"num_tokens": 2100665.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.533184623718261,
|
|
"epoch": 0.9118695873662761,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4851,
|
|
"mean_token_accuracy": 0.12483691051602364,
|
|
"num_tokens": 2112231.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.608133316040039,
|
|
"epoch": 0.9169638308711157,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4178,
|
|
"mean_token_accuracy": 0.1279417283833027,
|
|
"num_tokens": 2123920.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.607344055175782,
|
|
"epoch": 0.9220580743759552,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5409,
|
|
"mean_token_accuracy": 0.1146583005785942,
|
|
"num_tokens": 2135167.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.600939607620239,
|
|
"epoch": 0.9271523178807947,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5449,
|
|
"mean_token_accuracy": 0.11838355958461762,
|
|
"num_tokens": 2146809.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.6766938209533695,
|
|
"epoch": 0.9322465613856342,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4705,
|
|
"mean_token_accuracy": 0.1219531238079071,
|
|
"num_tokens": 2157784.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.527300453186035,
|
|
"epoch": 0.9373408048904738,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5532,
|
|
"mean_token_accuracy": 0.1199764221906662,
|
|
"num_tokens": 2169748.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.646005058288575,
|
|
"epoch": 0.9424350483953133,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5623,
|
|
"mean_token_accuracy": 0.11655192077159882,
|
|
"num_tokens": 2182370.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.536486101150513,
|
|
"epoch": 0.9475292919001528,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5354,
|
|
"mean_token_accuracy": 0.11724850311875343,
|
|
"num_tokens": 2196043.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.662149477005005,
|
|
"epoch": 0.9526235354049923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.5619,
|
|
"mean_token_accuracy": 0.11813377067446709,
|
|
"num_tokens": 2207240.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.638696384429932,
|
|
"epoch": 0.9577177789098319,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5672,
|
|
"mean_token_accuracy": 0.1213927574455738,
|
|
"num_tokens": 2219078.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.543146276473999,
|
|
"epoch": 0.9628120224146715,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5635,
|
|
"mean_token_accuracy": 0.1164170540869236,
|
|
"num_tokens": 2231434.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.707894611358642,
|
|
"epoch": 0.9679062659195109,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.476,
|
|
"mean_token_accuracy": 0.12743352800607682,
|
|
"num_tokens": 2241375.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.46073956489563,
|
|
"epoch": 0.9730005094243505,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5079,
|
|
"mean_token_accuracy": 0.12266820147633553,
|
|
"num_tokens": 2253520.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.511771488189697,
|
|
"epoch": 0.97809475292919,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4161,
|
|
"mean_token_accuracy": 0.12396993190050125,
|
|
"num_tokens": 2264938.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.533259010314941,
|
|
"epoch": 0.9831889964340296,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.5485,
|
|
"mean_token_accuracy": 0.11633272841572762,
|
|
"num_tokens": 2277306.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.535862684249878,
|
|
"epoch": 0.9882832399388691,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4161,
|
|
"mean_token_accuracy": 0.11685501486063003,
|
|
"num_tokens": 2288766.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.541786623001099,
|
|
"epoch": 0.9933774834437086,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4636,
|
|
"mean_token_accuracy": 0.12314928695559502,
|
|
"num_tokens": 2299487.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.465618991851807,
|
|
"epoch": 0.9984717269485481,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4102,
|
|
"mean_token_accuracy": 0.1320273019373417,
|
|
"num_tokens": 2311302.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.537135336134169,
|
|
"epoch": 1.0030565461029037,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3782,
|
|
"mean_token_accuracy": 0.12410805705520842,
|
|
"num_tokens": 2321410.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.459956550598145,
|
|
"epoch": 1.0081507896077433,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3087,
|
|
"mean_token_accuracy": 0.12975611090660094,
|
|
"num_tokens": 2332931.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.513260221481323,
|
|
"epoch": 1.0132450331125828,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3279,
|
|
"mean_token_accuracy": 0.12899856641888618,
|
|
"num_tokens": 2344207.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.3633050441741945,
|
|
"epoch": 1.0183392766174224,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2691,
|
|
"mean_token_accuracy": 0.13432745561003684,
|
|
"num_tokens": 2356470.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0183392766174224,
|
|
"eval_entropy": 6.235974288281099,
|
|
"eval_loss": 6.456569671630859,
|
|
"eval_mean_token_accuracy": 0.1273295690248042,
|
|
"eval_num_tokens": 2356470.0,
|
|
"eval_runtime": 2.7225,
|
|
"eval_samples_per_second": 1190.083,
|
|
"eval_steps_per_second": 148.76,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.592818260192871,
|
|
"epoch": 1.0234335201222617,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999997711134041,
|
|
"loss": 6.3144,
|
|
"mean_token_accuracy": 0.12813438922166825,
|
|
"num_tokens": 2368744.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.399446201324463,
|
|
"epoch": 1.0285277636271013,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999988412624068,
|
|
"loss": 6.3371,
|
|
"mean_token_accuracy": 0.12820167019963263,
|
|
"num_tokens": 2380906.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.4056432247161865,
|
|
"epoch": 1.0336220071319409,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999971961445492,
|
|
"loss": 6.2823,
|
|
"mean_token_accuracy": 0.13787578642368317,
|
|
"num_tokens": 2392358.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.404738092422486,
|
|
"epoch": 1.0387162506367804,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999948357650613,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.13328557312488556,
|
|
"num_tokens": 2405509.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.466694736480713,
|
|
"epoch": 1.04381049414162,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999917601314468,
|
|
"loss": 6.307,
|
|
"mean_token_accuracy": 0.12337398380041123,
|
|
"num_tokens": 2417890.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.453152513504028,
|
|
"epoch": 1.0489047376464595,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999879692534828,
|
|
"loss": 6.2815,
|
|
"mean_token_accuracy": 0.1301681473851204,
|
|
"num_tokens": 2429739.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.481996345520019,
|
|
"epoch": 1.053998981151299,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999834631432208,
|
|
"loss": 6.2261,
|
|
"mean_token_accuracy": 0.13167040944099426,
|
|
"num_tokens": 2441374.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.298508167266846,
|
|
"epoch": 1.0590932246561386,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999782418149852,
|
|
"loss": 6.2608,
|
|
"mean_token_accuracy": 0.12832411006093025,
|
|
"num_tokens": 2452914.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.416754484176636,
|
|
"epoch": 1.064187468160978,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999723052853747,
|
|
"loss": 6.3092,
|
|
"mean_token_accuracy": 0.1292288802564144,
|
|
"num_tokens": 2464970.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.4724352836608885,
|
|
"epoch": 1.0692817116658175,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999656535732615,
|
|
"loss": 6.3552,
|
|
"mean_token_accuracy": 0.12950649932026864,
|
|
"num_tokens": 2477563.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.425542736053467,
|
|
"epoch": 1.074375955170657,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999582866997913,
|
|
"loss": 6.2163,
|
|
"mean_token_accuracy": 0.1356003314256668,
|
|
"num_tokens": 2488767.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.325726699829102,
|
|
"epoch": 1.0794701986754967,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499950204688383,
|
|
"loss": 6.2706,
|
|
"mean_token_accuracy": 0.1314474955201149,
|
|
"num_tokens": 2500260.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.367005729675293,
|
|
"epoch": 1.0845644421803362,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999414075647292,
|
|
"loss": 6.2124,
|
|
"mean_token_accuracy": 0.13532492220401765,
|
|
"num_tokens": 2511581.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.4431493282318115,
|
|
"epoch": 1.0896586856851758,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499931895356796,
|
|
"loss": 6.3935,
|
|
"mean_token_accuracy": 0.12609082609415054,
|
|
"num_tokens": 2524185.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.36133770942688,
|
|
"epoch": 1.0947529291900153,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999216680948223,
|
|
"loss": 6.1905,
|
|
"mean_token_accuracy": 0.13383397907018663,
|
|
"num_tokens": 2535762.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.254091024398804,
|
|
"epoch": 1.0998471726948549,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999107258113204,
|
|
"loss": 6.1359,
|
|
"mean_token_accuracy": 0.14740005433559417,
|
|
"num_tokens": 2547566.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.3750659942626955,
|
|
"epoch": 1.1049414161996944,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998990685410759,
|
|
"loss": 6.3028,
|
|
"mean_token_accuracy": 0.12638821676373482,
|
|
"num_tokens": 2559796.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.344280099868774,
|
|
"epoch": 1.110035659704534,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998866963211465,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.13058622255921365,
|
|
"num_tokens": 2573137.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.395547533035279,
|
|
"epoch": 1.1151299032093733,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998736091908636,
|
|
"loss": 6.2859,
|
|
"mean_token_accuracy": 0.1297912612557411,
|
|
"num_tokens": 2586550.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.35224232673645,
|
|
"epoch": 1.1202241467142129,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998598071918309,
|
|
"loss": 6.3006,
|
|
"mean_token_accuracy": 0.13336251825094222,
|
|
"num_tokens": 2597666.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.355217361450196,
|
|
"epoch": 1.1253183902190524,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998452903679246,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.14084486067295074,
|
|
"num_tokens": 2608518.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.296790313720703,
|
|
"epoch": 1.130412633723892,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998300587652936,
|
|
"loss": 6.2659,
|
|
"mean_token_accuracy": 0.13195201009511948,
|
|
"num_tokens": 2619766.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.375978803634643,
|
|
"epoch": 1.1355068772287316,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998141124323586,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.13132200613617898,
|
|
"num_tokens": 2632473.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.365690469741821,
|
|
"epoch": 1.1406011207335711,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997974514198126,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.13221723809838296,
|
|
"num_tokens": 2644059.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.3026340961456295,
|
|
"epoch": 1.1456953642384107,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997800757806209,
|
|
"loss": 6.194,
|
|
"mean_token_accuracy": 0.1333719238638878,
|
|
"num_tokens": 2655168.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.4999518394470215,
|
|
"epoch": 1.15078960774325,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997619855700203,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.1288960285484791,
|
|
"num_tokens": 2665431.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.237315464019775,
|
|
"epoch": 1.1558838512480896,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499743180845519,
|
|
"loss": 6.2318,
|
|
"mean_token_accuracy": 0.1425723135471344,
|
|
"num_tokens": 2676942.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.455456638336182,
|
|
"epoch": 1.1609780947529291,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499723661666897,
|
|
"loss": 6.292,
|
|
"mean_token_accuracy": 0.1302104614675045,
|
|
"num_tokens": 2688510.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.28514461517334,
|
|
"epoch": 1.1660723382577687,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997034280962054,
|
|
"loss": 6.2872,
|
|
"mean_token_accuracy": 0.13086067363619805,
|
|
"num_tokens": 2702140.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.3039695739746096,
|
|
"epoch": 1.1711665817626082,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996824801977663,
|
|
"loss": 6.2605,
|
|
"mean_token_accuracy": 0.1339351736009121,
|
|
"num_tokens": 2714358.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.298797225952148,
|
|
"epoch": 1.1762608252674478,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996608180381727,
|
|
"loss": 6.1893,
|
|
"mean_token_accuracy": 0.13878541216254234,
|
|
"num_tokens": 2725477.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.347237777709961,
|
|
"epoch": 1.1813550687722874,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499638441686288,
|
|
"loss": 6.3059,
|
|
"mean_token_accuracy": 0.12969013229012488,
|
|
"num_tokens": 2736696.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.369765949249268,
|
|
"epoch": 1.186449312277127,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996153512132462,
|
|
"loss": 6.1794,
|
|
"mean_token_accuracy": 0.14288636445999145,
|
|
"num_tokens": 2747655.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.2775472640991214,
|
|
"epoch": 1.1915435557819665,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995915466924517,
|
|
"loss": 6.2175,
|
|
"mean_token_accuracy": 0.1385953515768051,
|
|
"num_tokens": 2758955.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.286406564712524,
|
|
"epoch": 1.196637799286806,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995670281995784,
|
|
"loss": 6.1473,
|
|
"mean_token_accuracy": 0.14105831906199456,
|
|
"num_tokens": 2770395.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.273731756210327,
|
|
"epoch": 1.2017320427916454,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004995417958125703,
|
|
"loss": 6.2511,
|
|
"mean_token_accuracy": 0.13191839084029197,
|
|
"num_tokens": 2783043.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.376677417755127,
|
|
"epoch": 1.206826286296485,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995158496116402,
|
|
"loss": 6.372,
|
|
"mean_token_accuracy": 0.12635432854294776,
|
|
"num_tokens": 2795937.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.3480470180511475,
|
|
"epoch": 1.2119205298013245,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004994891896792712,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.13276322185993195,
|
|
"num_tokens": 2807812.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.292226409912109,
|
|
"epoch": 1.217014773306164,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004994618161002142,
|
|
"loss": 6.2304,
|
|
"mean_token_accuracy": 0.13926912918686868,
|
|
"num_tokens": 2819326.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.294836330413818,
|
|
"epoch": 1.2221090168110036,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994337289614893,
|
|
"loss": 6.1696,
|
|
"mean_token_accuracy": 0.1429339349269867,
|
|
"num_tokens": 2830917.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.302328395843506,
|
|
"epoch": 1.2272032603158431,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994049283523854,
|
|
"loss": 6.2782,
|
|
"mean_token_accuracy": 0.134336144477129,
|
|
"num_tokens": 2842828.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.213818788528442,
|
|
"epoch": 1.2322975038206827,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993754143644585,
|
|
"loss": 6.0664,
|
|
"mean_token_accuracy": 0.14119207188487054,
|
|
"num_tokens": 2854423.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.279035377502441,
|
|
"epoch": 1.237391747325522,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004993451870915333,
|
|
"loss": 6.2969,
|
|
"mean_token_accuracy": 0.13581560626626016,
|
|
"num_tokens": 2866032.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.2984619140625,
|
|
"epoch": 1.2424859908303616,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993142466297016,
|
|
"loss": 6.2562,
|
|
"mean_token_accuracy": 0.13544865399599076,
|
|
"num_tokens": 2877147.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.162362670898437,
|
|
"epoch": 1.2475802343352012,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004992825930773228,
|
|
"loss": 6.156,
|
|
"mean_token_accuracy": 0.14262611120939256,
|
|
"num_tokens": 2888723.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.353058528900147,
|
|
"epoch": 1.2526744778400407,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992502265350224,
|
|
"loss": 6.2198,
|
|
"mean_token_accuracy": 0.1357116438448429,
|
|
"num_tokens": 2901158.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.2556548595428465,
|
|
"epoch": 1.2577687213448803,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004992171471056933,
|
|
"loss": 6.236,
|
|
"mean_token_accuracy": 0.13508779555559158,
|
|
"num_tokens": 2913303.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.30905237197876,
|
|
"epoch": 1.2628629648497198,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004991833548944945,
|
|
"loss": 6.2052,
|
|
"mean_token_accuracy": 0.13921287879347802,
|
|
"num_tokens": 2924036.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.259322547912598,
|
|
"epoch": 1.2679572083545594,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004991488500088505,
|
|
"loss": 6.3012,
|
|
"mean_token_accuracy": 0.1355997823178768,
|
|
"num_tokens": 2936031.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.29479570388794,
|
|
"epoch": 1.273051451859399,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991136325584516,
|
|
"loss": 6.1903,
|
|
"mean_token_accuracy": 0.13741473108530045,
|
|
"num_tokens": 2947707.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.217977571487427,
|
|
"epoch": 1.2781456953642385,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990777026552536,
|
|
"loss": 6.1573,
|
|
"mean_token_accuracy": 0.13459015190601348,
|
|
"num_tokens": 2959212.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.257912874221802,
|
|
"epoch": 1.283239938869078,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990410604134769,
|
|
"loss": 6.1617,
|
|
"mean_token_accuracy": 0.1383515976369381,
|
|
"num_tokens": 2969323.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.30942325592041,
|
|
"epoch": 1.2883341823739174,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004990037059496064,
|
|
"loss": 6.2848,
|
|
"mean_token_accuracy": 0.13528576493263245,
|
|
"num_tokens": 2981194.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.265507745742798,
|
|
"epoch": 1.293428425878757,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004989656393823913,
|
|
"loss": 6.1875,
|
|
"mean_token_accuracy": 0.13480680510401727,
|
|
"num_tokens": 2992373.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.194509410858155,
|
|
"epoch": 1.2985226693835965,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004989268608328443,
|
|
"loss": 6.1929,
|
|
"mean_token_accuracy": 0.13894519731402397,
|
|
"num_tokens": 3004019.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.3316387176513675,
|
|
"epoch": 1.303616912888436,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004988873704242418,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.13455382362008095,
|
|
"num_tokens": 3016393.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.134871530532837,
|
|
"epoch": 1.3087111563932756,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004988471682821229,
|
|
"loss": 6.1498,
|
|
"mean_token_accuracy": 0.13220743089914322,
|
|
"num_tokens": 3027062.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.240565252304077,
|
|
"epoch": 1.3138053998981152,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004988062545342894,
|
|
"loss": 6.1463,
|
|
"mean_token_accuracy": 0.13820229694247246,
|
|
"num_tokens": 3039500.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.211811256408692,
|
|
"epoch": 1.3188996434029547,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004987646293108056,
|
|
"loss": 6.1094,
|
|
"mean_token_accuracy": 0.143786983191967,
|
|
"num_tokens": 3049296.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.245458555221558,
|
|
"epoch": 1.323993886907794,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987222927439967,
|
|
"loss": 6.1159,
|
|
"mean_token_accuracy": 0.14654901921749114,
|
|
"num_tokens": 3059806.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.349328660964966,
|
|
"epoch": 1.3290881304126336,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004986792449684504,
|
|
"loss": 6.2837,
|
|
"mean_token_accuracy": 0.13811362013220788,
|
|
"num_tokens": 3072769.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.191184425354004,
|
|
"epoch": 1.3341823739174732,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004986354861210144,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.1372109092772007,
|
|
"num_tokens": 3084547.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.320459651947021,
|
|
"epoch": 1.3392766174223127,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004985910163407973,
|
|
"loss": 6.2426,
|
|
"mean_token_accuracy": 0.1373012289404869,
|
|
"num_tokens": 3095905.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.322480058670044,
|
|
"epoch": 1.3443708609271523,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985458357691675,
|
|
"loss": 6.2376,
|
|
"mean_token_accuracy": 0.13577743023633956,
|
|
"num_tokens": 3106981.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.263123798370361,
|
|
"epoch": 1.3494651044319919,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004984999445497533,
|
|
"loss": 6.2013,
|
|
"mean_token_accuracy": 0.14307116642594336,
|
|
"num_tokens": 3117916.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.290118980407715,
|
|
"epoch": 1.3545593479368314,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000498453342828442,
|
|
"loss": 6.2349,
|
|
"mean_token_accuracy": 0.1347646377980709,
|
|
"num_tokens": 3129826.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.295152711868286,
|
|
"epoch": 1.359653591441671,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004984060307533794,
|
|
"loss": 6.3028,
|
|
"mean_token_accuracy": 0.13658893331885338,
|
|
"num_tokens": 3142543.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.158872127532959,
|
|
"epoch": 1.3647478349465105,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004983580084749698,
|
|
"loss": 6.1885,
|
|
"mean_token_accuracy": 0.1535182163119316,
|
|
"num_tokens": 3154192.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.224641418457031,
|
|
"epoch": 1.36984207845135,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498309276145875,
|
|
"loss": 6.1083,
|
|
"mean_token_accuracy": 0.14207091629505159,
|
|
"num_tokens": 3166495.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.165792417526245,
|
|
"epoch": 1.3749363219561896,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004982598339210142,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.1307465046644211,
|
|
"num_tokens": 3177798.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.343322038650513,
|
|
"epoch": 1.380030565461029,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004982096819575631,
|
|
"loss": 6.1695,
|
|
"mean_token_accuracy": 0.14079055488109588,
|
|
"num_tokens": 3189625.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.116537761688233,
|
|
"epoch": 1.3851248089658685,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981588204149538,
|
|
"loss": 6.1561,
|
|
"mean_token_accuracy": 0.14167958348989487,
|
|
"num_tokens": 3202247.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.212584590911865,
|
|
"epoch": 1.390219052470708,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004981072494548743,
|
|
"loss": 6.1181,
|
|
"mean_token_accuracy": 0.14052650555968285,
|
|
"num_tokens": 3214566.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.271206998825074,
|
|
"epoch": 1.3953132959755477,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004980549692412676,
|
|
"loss": 6.2696,
|
|
"mean_token_accuracy": 0.13090315014123916,
|
|
"num_tokens": 3228692.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.249164581298828,
|
|
"epoch": 1.4004075394803872,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004980019799403313,
|
|
"loss": 6.2175,
|
|
"mean_token_accuracy": 0.13499587401747704,
|
|
"num_tokens": 3240933.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.0497523784637455,
|
|
"epoch": 1.4055017829852268,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004979482817205176,
|
|
"loss": 6.0231,
|
|
"mean_token_accuracy": 0.1431894138455391,
|
|
"num_tokens": 3252007.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.247428750991821,
|
|
"epoch": 1.410596026490066,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004978938747525319,
|
|
"loss": 6.1064,
|
|
"mean_token_accuracy": 0.14167555868625642,
|
|
"num_tokens": 3262976.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.091307783126831,
|
|
"epoch": 1.4156902699949057,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004978387592093329,
|
|
"loss": 6.0763,
|
|
"mean_token_accuracy": 0.13792687356472016,
|
|
"num_tokens": 3274352.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.242980241775513,
|
|
"epoch": 1.4207845134997452,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004977829352661318,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.14062171727418898,
|
|
"num_tokens": 3287615.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.209613513946533,
|
|
"epoch": 1.4258787570045848,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004977264031003919,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.1420186996459961,
|
|
"num_tokens": 3300086.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.068993854522705,
|
|
"epoch": 1.4309730005094243,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976691628918278,
|
|
"loss": 6.0239,
|
|
"mean_token_accuracy": 0.14449454843997955,
|
|
"num_tokens": 3310961.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.247696733474731,
|
|
"epoch": 1.4360672440142639,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000497611214822405,
|
|
"loss": 6.1349,
|
|
"mean_token_accuracy": 0.1391749322414398,
|
|
"num_tokens": 3322077.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.161630392074585,
|
|
"epoch": 1.4411614875191034,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004975525590763393,
|
|
"loss": 6.1601,
|
|
"mean_token_accuracy": 0.1388368107378483,
|
|
"num_tokens": 3332725.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.150338840484619,
|
|
"epoch": 1.446255731023943,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497493195840096,
|
|
"loss": 6.1938,
|
|
"mean_token_accuracy": 0.13313170447945594,
|
|
"num_tokens": 3344267.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.278788948059082,
|
|
"epoch": 1.4513499745287826,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974331253023898,
|
|
"loss": 6.2198,
|
|
"mean_token_accuracy": 0.1363731674849987,
|
|
"num_tokens": 3356947.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.222986936569214,
|
|
"epoch": 1.4564442180336221,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004973723476541836,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.14745828732848168,
|
|
"num_tokens": 3368934.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.136757898330688,
|
|
"epoch": 1.4615384615384617,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004973108630886885,
|
|
"loss": 6.2628,
|
|
"mean_token_accuracy": 0.13224418684840203,
|
|
"num_tokens": 3380806.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.247390556335449,
|
|
"epoch": 1.466632705043301,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004972486718013626,
|
|
"loss": 6.1936,
|
|
"mean_token_accuracy": 0.14039151445031167,
|
|
"num_tokens": 3393469.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.144114685058594,
|
|
"epoch": 1.4717269485481406,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004971857739899109,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.14480106085538863,
|
|
"num_tokens": 3405667.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.206645774841308,
|
|
"epoch": 1.4768211920529801,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004971221698542842,
|
|
"loss": 6.2177,
|
|
"mean_token_accuracy": 0.13361408412456513,
|
|
"num_tokens": 3417477.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.156244373321533,
|
|
"epoch": 1.4819154355578197,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004970578595966789,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.14549017548561097,
|
|
"num_tokens": 3429395.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.194065904617309,
|
|
"epoch": 1.4870096790626592,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969928434215358,
|
|
"loss": 6.1722,
|
|
"mean_token_accuracy": 0.14258456006646156,
|
|
"num_tokens": 3441135.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.209768962860108,
|
|
"epoch": 1.4921039225674988,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969271215355402,
|
|
"loss": 6.2036,
|
|
"mean_token_accuracy": 0.13492640405893325,
|
|
"num_tokens": 3453712.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.254387140274048,
|
|
"epoch": 1.4971981660723381,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004968606941476207,
|
|
"loss": 6.1736,
|
|
"mean_token_accuracy": 0.13807964622974395,
|
|
"num_tokens": 3466346.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.154835987091064,
|
|
"epoch": 1.5022924095771777,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004967935614689485,
|
|
"loss": 6.137,
|
|
"mean_token_accuracy": 0.14218896627426147,
|
|
"num_tokens": 3478393.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.075071382522583,
|
|
"epoch": 1.5073866530820172,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004967257237129373,
|
|
"loss": 6.0347,
|
|
"mean_token_accuracy": 0.15178611278533935,
|
|
"num_tokens": 3489631.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.193858480453491,
|
|
"epoch": 1.5124808965868568,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004966571810952417,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.13910189419984817,
|
|
"num_tokens": 3502403.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.017774820327759,
|
|
"epoch": 1.5175751400916964,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004965879338337577,
|
|
"loss": 6.0344,
|
|
"mean_token_accuracy": 0.15021622478961943,
|
|
"num_tokens": 3513512.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.203876113891601,
|
|
"epoch": 1.522669383596536,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004965179821486205,
|
|
"loss": 6.2174,
|
|
"mean_token_accuracy": 0.13455724492669105,
|
|
"num_tokens": 3525670.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.092337226867675,
|
|
"epoch": 1.5277636271013755,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004964473262622054,
|
|
"loss": 6.1109,
|
|
"mean_token_accuracy": 0.13562070727348327,
|
|
"num_tokens": 3537634.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.12050495147705,
|
|
"epoch": 1.532857870606215,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496375966399126,
|
|
"loss": 6.0473,
|
|
"mean_token_accuracy": 0.1405868835747242,
|
|
"num_tokens": 3549344.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.208581924438477,
|
|
"epoch": 1.5379521141110546,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000496303902786234,
|
|
"loss": 6.1977,
|
|
"mean_token_accuracy": 0.14212006479501724,
|
|
"num_tokens": 3563193.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.079247188568115,
|
|
"epoch": 1.5430463576158941,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004962311356526181,
|
|
"loss": 6.1229,
|
|
"mean_token_accuracy": 0.14019954353570938,
|
|
"num_tokens": 3574304.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.252691125869751,
|
|
"epoch": 1.5481406011207337,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004961576652296036,
|
|
"loss": 6.0225,
|
|
"mean_token_accuracy": 0.1485082432627678,
|
|
"num_tokens": 3586396.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.026348543167114,
|
|
"epoch": 1.5532348446255733,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004960834917507516,
|
|
"loss": 6.0247,
|
|
"mean_token_accuracy": 0.1489784114062786,
|
|
"num_tokens": 3597945.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.194207620620728,
|
|
"epoch": 1.5583290881304126,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004960086154518578,
|
|
"loss": 6.128,
|
|
"mean_token_accuracy": 0.1477145791053772,
|
|
"num_tokens": 3610378.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.122859382629395,
|
|
"epoch": 1.5634233316352522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004959330365709528,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.14782411605119705,
|
|
"num_tokens": 3621768.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.11494517326355,
|
|
"epoch": 1.5685175751400917,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004958567553483003,
|
|
"loss": 6.0833,
|
|
"mean_token_accuracy": 0.14180212020874022,
|
|
"num_tokens": 3634473.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.082244253158569,
|
|
"epoch": 1.5736118186449313,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004957797720263967,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.1548992156982422,
|
|
"num_tokens": 3644499.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.092272138595581,
|
|
"epoch": 1.5787060621497706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004957020868499706,
|
|
"loss": 6.0399,
|
|
"mean_token_accuracy": 0.14433746933937072,
|
|
"num_tokens": 3655807.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.138490962982178,
|
|
"epoch": 1.5838003056546102,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004956237000659815,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.15379355400800704,
|
|
"num_tokens": 3667301.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.990786409378051,
|
|
"epoch": 1.5888945491594497,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004955446119236195,
|
|
"loss": 6.1016,
|
|
"mean_token_accuracy": 0.14067147970199584,
|
|
"num_tokens": 3679448.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.1579066753387455,
|
|
"epoch": 1.5939887926642893,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004954648226743043,
|
|
"loss": 6.0538,
|
|
"mean_token_accuracy": 0.1469771906733513,
|
|
"num_tokens": 3690985.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.109905433654785,
|
|
"epoch": 1.5990830361691288,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004953843325716841,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.14429525285959244,
|
|
"num_tokens": 3703677.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.18586049079895,
|
|
"epoch": 1.6041772796739684,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004953031418716356,
|
|
"loss": 6.0519,
|
|
"mean_token_accuracy": 0.15168690159916878,
|
|
"num_tokens": 3715429.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.018005847930908,
|
|
"epoch": 1.609271523178808,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004952212508322623,
|
|
"loss": 6.1098,
|
|
"mean_token_accuracy": 0.14498841464519502,
|
|
"num_tokens": 3727663.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.126214218139649,
|
|
"epoch": 1.6143657666836475,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004951386597138943,
|
|
"loss": 6.0257,
|
|
"mean_token_accuracy": 0.14750143885612488,
|
|
"num_tokens": 3739585.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.134209012985229,
|
|
"epoch": 1.619460010188487,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004950553687790871,
|
|
"loss": 6.167,
|
|
"mean_token_accuracy": 0.14012180045247077,
|
|
"num_tokens": 3751382.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.058493661880493,
|
|
"epoch": 1.6245542536933266,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004949713782926208,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.15354147255420686,
|
|
"num_tokens": 3762520.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.086709356307983,
|
|
"epoch": 1.6296484971981662,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004948866885214996,
|
|
"loss": 6.0176,
|
|
"mean_token_accuracy": 0.14750826209783555,
|
|
"num_tokens": 3773821.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.167296695709228,
|
|
"epoch": 1.6347427407030057,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004948012997349506,
|
|
"loss": 6.095,
|
|
"mean_token_accuracy": 0.1464608058333397,
|
|
"num_tokens": 3785767.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.106127071380615,
|
|
"epoch": 1.6398369842078453,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004947152122044232,
|
|
"loss": 6.1243,
|
|
"mean_token_accuracy": 0.14179719388484954,
|
|
"num_tokens": 3796702.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.088868474960327,
|
|
"epoch": 1.6449312277126846,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004946284262035875,
|
|
"loss": 6.0353,
|
|
"mean_token_accuracy": 0.14263806939125062,
|
|
"num_tokens": 3808757.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.1169596195220945,
|
|
"epoch": 1.6500254712175242,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000494540942008335,
|
|
"loss": 6.0118,
|
|
"mean_token_accuracy": 0.150656595826149,
|
|
"num_tokens": 3819804.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.100694417953491,
|
|
"epoch": 1.6551197147223637,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000494452759896776,
|
|
"loss": 6.0563,
|
|
"mean_token_accuracy": 0.14616145342588424,
|
|
"num_tokens": 3831877.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.152666425704956,
|
|
"epoch": 1.6602139582272033,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943638801492396,
|
|
"loss": 6.0227,
|
|
"mean_token_accuracy": 0.14526450335979463,
|
|
"num_tokens": 3843356.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.038093042373657,
|
|
"epoch": 1.6653082017320426,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004942743030482728,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.14405328556895255,
|
|
"num_tokens": 3854779.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.060023975372315,
|
|
"epoch": 1.6704024452368822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004941840288786397,
|
|
"loss": 6.0095,
|
|
"mean_token_accuracy": 0.1491662733256817,
|
|
"num_tokens": 3867309.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.107427787780762,
|
|
"epoch": 1.6754966887417218,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004940930579273198,
|
|
"loss": 6.0939,
|
|
"mean_token_accuracy": 0.14735165163874625,
|
|
"num_tokens": 3879341.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.110239028930664,
|
|
"epoch": 1.6805909322465613,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004940013904835079,
|
|
"loss": 5.9833,
|
|
"mean_token_accuracy": 0.1522666648030281,
|
|
"num_tokens": 3890639.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.052538728713989,
|
|
"epoch": 1.6856851757514009,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004939090268386132,
|
|
"loss": 6.0413,
|
|
"mean_token_accuracy": 0.14310309216380118,
|
|
"num_tokens": 3904185.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.085521507263183,
|
|
"epoch": 1.6907794192562404,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004938159672862578,
|
|
"loss": 6.0194,
|
|
"mean_token_accuracy": 0.14190833568572997,
|
|
"num_tokens": 3917137.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.077161645889282,
|
|
"epoch": 1.69587366276108,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004937222121222763,
|
|
"loss": 6.0667,
|
|
"mean_token_accuracy": 0.14442135989665986,
|
|
"num_tokens": 3929911.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.157588338851928,
|
|
"epoch": 1.7009679062659195,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004936277616447144,
|
|
"loss": 6.0951,
|
|
"mean_token_accuracy": 0.1374649502336979,
|
|
"num_tokens": 3941242.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.070014762878418,
|
|
"epoch": 1.706062149770759,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004935326161538281,
|
|
"loss": 6.0615,
|
|
"mean_token_accuracy": 0.1456019811332226,
|
|
"num_tokens": 3952464.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.071111249923706,
|
|
"epoch": 1.7111563932755987,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004934367759520834,
|
|
"loss": 6.0197,
|
|
"mean_token_accuracy": 0.14644915908575057,
|
|
"num_tokens": 3965479.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.0241259098052975,
|
|
"epoch": 1.7162506367804382,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004933402413441541,
|
|
"loss": 6.0303,
|
|
"mean_token_accuracy": 0.14970999360084533,
|
|
"num_tokens": 3978195.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.169180154800415,
|
|
"epoch": 1.7213448802852778,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004932430126369221,
|
|
"loss": 6.1285,
|
|
"mean_token_accuracy": 0.13854920864105225,
|
|
"num_tokens": 3989373.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.978077363967896,
|
|
"epoch": 1.7264391237901173,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004931450901394753,
|
|
"loss": 6.0419,
|
|
"mean_token_accuracy": 0.14705563187599183,
|
|
"num_tokens": 4000860.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.1055442810058596,
|
|
"epoch": 1.7315333672949567,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004930464741631073,
|
|
"loss": 6.0047,
|
|
"mean_token_accuracy": 0.14636100381612777,
|
|
"num_tokens": 4012035.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.03545560836792,
|
|
"epoch": 1.7366276107997962,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004929471650213167,
|
|
"loss": 6.0092,
|
|
"mean_token_accuracy": 0.149752613902092,
|
|
"num_tokens": 4024490.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.236439847946167,
|
|
"epoch": 1.7417218543046358,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004928471630298051,
|
|
"loss": 6.1215,
|
|
"mean_token_accuracy": 0.14440922439098358,
|
|
"num_tokens": 4036277.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.903345203399658,
|
|
"epoch": 1.7468160978094753,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004927464685064767,
|
|
"loss": 5.9182,
|
|
"mean_token_accuracy": 0.1580268085002899,
|
|
"num_tokens": 4047734.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.989237260818482,
|
|
"epoch": 1.7519103413143147,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004926450817714376,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.15351947844028474,
|
|
"num_tokens": 4059037.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.020188474655152,
|
|
"epoch": 1.7570045848191542,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004925430031469941,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.1485172614455223,
|
|
"num_tokens": 4070795.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.062839460372925,
|
|
"epoch": 1.7620988283239938,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004924402329576522,
|
|
"loss": 5.8529,
|
|
"mean_token_accuracy": 0.1641454890370369,
|
|
"num_tokens": 4082661.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.956881666183472,
|
|
"epoch": 1.7671930718288333,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004923367715301164,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.14976243525743485,
|
|
"num_tokens": 4093151.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.009133148193359,
|
|
"epoch": 1.772287315333673,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004922326191932882,
|
|
"loss": 6.0323,
|
|
"mean_token_accuracy": 0.14575401470065116,
|
|
"num_tokens": 4104380.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.06882758140564,
|
|
"epoch": 1.7773815588385125,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004921277762782661,
|
|
"loss": 5.9418,
|
|
"mean_token_accuracy": 0.1585811659693718,
|
|
"num_tokens": 4115506.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.0451396942138675,
|
|
"epoch": 1.782475802343352,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004920222431183437,
|
|
"loss": 6.0799,
|
|
"mean_token_accuracy": 0.14764982908964158,
|
|
"num_tokens": 4126978.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.010602045059204,
|
|
"epoch": 1.7875700458481916,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919160200490085,
|
|
"loss": 5.9057,
|
|
"mean_token_accuracy": 0.1553368128836155,
|
|
"num_tokens": 4137602.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.043928575515747,
|
|
"epoch": 1.7926642893530311,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004918091074079419,
|
|
"loss": 6.005,
|
|
"mean_token_accuracy": 0.14657752737402915,
|
|
"num_tokens": 4148617.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.101671123504639,
|
|
"epoch": 1.7977585328578707,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004917015055350167,
|
|
"loss": 6.1125,
|
|
"mean_token_accuracy": 0.1507917121052742,
|
|
"num_tokens": 4162445.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.156926488876342,
|
|
"epoch": 1.8028527763627102,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004915932147722974,
|
|
"loss": 6.088,
|
|
"mean_token_accuracy": 0.1469472289085388,
|
|
"num_tokens": 4175163.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.04713454246521,
|
|
"epoch": 1.8079470198675498,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004914842354640382,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.14839993864297868,
|
|
"num_tokens": 4189159.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.123418474197388,
|
|
"epoch": 1.8130412633723894,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491374567956682,
|
|
"loss": 6.151,
|
|
"mean_token_accuracy": 0.13950974345207215,
|
|
"num_tokens": 4201989.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.006437921524048,
|
|
"epoch": 1.8181355068772287,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004912642125988597,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.154378542304039,
|
|
"num_tokens": 4214982.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.0241734981536865,
|
|
"epoch": 1.8232297503820682,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004911531697413888,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.15884421169757842,
|
|
"num_tokens": 4225932.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.019749498367309,
|
|
"epoch": 1.8283239938869078,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004910414397372724,
|
|
"loss": 6.0673,
|
|
"mean_token_accuracy": 0.1400178052484989,
|
|
"num_tokens": 4238406.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.000314950942993,
|
|
"epoch": 1.8334182373917474,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004909290229416979,
|
|
"loss": 5.8763,
|
|
"mean_token_accuracy": 0.1577962964773178,
|
|
"num_tokens": 4249956.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.961596298217773,
|
|
"epoch": 1.8385124808965867,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490815919712036,
|
|
"loss": 6.0199,
|
|
"mean_token_accuracy": 0.1445288985967636,
|
|
"num_tokens": 4262367.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.051313352584839,
|
|
"epoch": 1.8436067244014263,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004907021304078398,
|
|
"loss": 6.0205,
|
|
"mean_token_accuracy": 0.1489431828260422,
|
|
"num_tokens": 4273767.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.014533138275146,
|
|
"epoch": 1.8487009679062658,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004905876553908432,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.15442364513874055,
|
|
"num_tokens": 4285545.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.138439273834228,
|
|
"epoch": 1.8537952114111054,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004904724950249599,
|
|
"loss": 6.032,
|
|
"mean_token_accuracy": 0.1408793196082115,
|
|
"num_tokens": 4296963.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.9494537830352785,
|
|
"epoch": 1.858889454915945,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004903566496762824,
|
|
"loss": 5.8881,
|
|
"mean_token_accuracy": 0.15565152913331987,
|
|
"num_tokens": 4307724.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.986615371704102,
|
|
"epoch": 1.8639836984207845,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490240119713081,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.15582243353128433,
|
|
"num_tokens": 4319153.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.897438669204712,
|
|
"epoch": 1.869077941925624,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004901229055058019,
|
|
"loss": 5.9029,
|
|
"mean_token_accuracy": 0.15601452589035034,
|
|
"num_tokens": 4330179.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.983881711959839,
|
|
"epoch": 1.8741721854304636,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004900050074270669,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.1523781821131706,
|
|
"num_tokens": 4340747.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.045437383651733,
|
|
"epoch": 1.8792664289353032,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004898864258516717,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.14963105991482734,
|
|
"num_tokens": 4352336.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.010454797744751,
|
|
"epoch": 1.8843606724401427,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004897671611565845,
|
|
"loss": 5.8855,
|
|
"mean_token_accuracy": 0.15026046931743622,
|
|
"num_tokens": 4363189.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.973527097702027,
|
|
"epoch": 1.8894549159449823,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004896472137209457,
|
|
"loss": 6.0435,
|
|
"mean_token_accuracy": 0.15217291116714476,
|
|
"num_tokens": 4375946.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.047983837127686,
|
|
"epoch": 1.8945491594498218,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004895265839260657,
|
|
"loss": 5.9874,
|
|
"mean_token_accuracy": 0.15549195781350136,
|
|
"num_tokens": 4387543.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.011504077911377,
|
|
"epoch": 1.8996434029546614,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004894052721554242,
|
|
"loss": 6.0133,
|
|
"mean_token_accuracy": 0.15051916539669036,
|
|
"num_tokens": 4399790.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.112524843215942,
|
|
"epoch": 1.9047376464595007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004892832787946691,
|
|
"loss": 6.0419,
|
|
"mean_token_accuracy": 0.1465665489435196,
|
|
"num_tokens": 4411321.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.99403190612793,
|
|
"epoch": 1.9098318899643403,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004891606042316146,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.15396562218666077,
|
|
"num_tokens": 4422712.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.019381666183472,
|
|
"epoch": 1.9149261334691798,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004890372488562408,
|
|
"loss": 5.9759,
|
|
"mean_token_accuracy": 0.14543071910738944,
|
|
"num_tokens": 4434941.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.02582426071167,
|
|
"epoch": 1.9200203769740194,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004889132130606921,
|
|
"loss": 6.0101,
|
|
"mean_token_accuracy": 0.15340268537402152,
|
|
"num_tokens": 4447440.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.005884313583374,
|
|
"epoch": 1.9251146204788587,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004887884972392758,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.148632150888443,
|
|
"num_tokens": 4459086.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.995988559722901,
|
|
"epoch": 1.9302088639836983,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000488663101788461,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.1570799633860588,
|
|
"num_tokens": 4470463.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.92781023979187,
|
|
"epoch": 1.9353031074885378,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004885370271068773,
|
|
"loss": 5.9379,
|
|
"mean_token_accuracy": 0.1514614850282669,
|
|
"num_tokens": 4481266.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.934862375259399,
|
|
"epoch": 1.9403973509933774,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004884102735953136,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.15483575761318208,
|
|
"num_tokens": 4493367.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.038597154617309,
|
|
"epoch": 1.945491594498217,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000488282841656717,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.1498335026204586,
|
|
"num_tokens": 4505317.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.953360557556152,
|
|
"epoch": 1.9505858380030565,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048815473169619097,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.1488858863711357,
|
|
"num_tokens": 4516743.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.039535713195801,
|
|
"epoch": 1.955680081507896,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004880259441209945,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.15073336511850358,
|
|
"num_tokens": 4529467.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.9129636764526365,
|
|
"epoch": 1.9607743250127356,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004878964793405408,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.16155888140201569,
|
|
"num_tokens": 4540648.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.907260608673096,
|
|
"epoch": 1.9658685685175752,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048776633776639584,
|
|
"loss": 5.8924,
|
|
"mean_token_accuracy": 0.15700554400682448,
|
|
"num_tokens": 4552095.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.97724757194519,
|
|
"epoch": 1.9709628120224147,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048763551981227703,
|
|
"loss": 5.9128,
|
|
"mean_token_accuracy": 0.15258407592773438,
|
|
"num_tokens": 4563286.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.059520292282104,
|
|
"epoch": 1.9760570555272543,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048750402589405224,
|
|
"loss": 5.9816,
|
|
"mean_token_accuracy": 0.154987071454525,
|
|
"num_tokens": 4575313.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.934057950973511,
|
|
"epoch": 1.9811512990320939,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048737185642973784,
|
|
"loss": 5.9296,
|
|
"mean_token_accuracy": 0.15229681879281998,
|
|
"num_tokens": 4587155.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.014066076278686,
|
|
"epoch": 1.9862455425369334,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487239011839498,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.1469045899808407,
|
|
"num_tokens": 4598176.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.044819021224976,
|
|
"epoch": 1.9913397860417728,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004871054925456431,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.1422971233725548,
|
|
"num_tokens": 4610004.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.008854770660401,
|
|
"epoch": 1.9964340295466123,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048697129897262833,
|
|
"loss": 5.9429,
|
|
"mean_token_accuracy": 0.15738821029663086,
|
|
"num_tokens": 4620985.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.906723870171441,
|
|
"epoch": 2.001018848700968,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004868364315470523,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.1615782810582055,
|
|
"num_tokens": 4630550.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.944497156143188,
|
|
"epoch": 2.0061130922058075,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048670089069765594,
|
|
"loss": 5.5724,
|
|
"mean_token_accuracy": 0.1649057984352112,
|
|
"num_tokens": 4642173.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.895261764526367,
|
|
"epoch": 2.011207335710647,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004865646768553209,
|
|
"loss": 5.5952,
|
|
"mean_token_accuracy": 0.17154349386692047,
|
|
"num_tokens": 4654322.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.866156148910522,
|
|
"epoch": 2.0163015792154866,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004864277904530684,
|
|
"loss": 5.5887,
|
|
"mean_token_accuracy": 0.1739685967564583,
|
|
"num_tokens": 4665716.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.929111099243164,
|
|
"epoch": 2.021395822720326,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048629023192605747,
|
|
"loss": 5.6832,
|
|
"mean_token_accuracy": 0.17015760242938996,
|
|
"num_tokens": 4677148.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.944976711273194,
|
|
"epoch": 2.0264900662251657,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048615200171158395,
|
|
"loss": 5.6005,
|
|
"mean_token_accuracy": 0.16912046670913697,
|
|
"num_tokens": 4690855.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.803352785110474,
|
|
"epoch": 2.0315843097300053,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048601310024907904,
|
|
"loss": 5.5836,
|
|
"mean_token_accuracy": 0.16719031259417533,
|
|
"num_tokens": 4702933.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.921322536468506,
|
|
"epoch": 2.036678553234845,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048587352798010766,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.15699441730976105,
|
|
"num_tokens": 4715584.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.036678553234845,
|
|
"eval_entropy": 5.672201461556517,
|
|
"eval_loss": 5.9956374168396,
|
|
"eval_mean_token_accuracy": 0.1547820175503507,
|
|
"eval_num_tokens": 4715584.0,
|
|
"eval_runtime": 2.9019,
|
|
"eval_samples_per_second": 1116.523,
|
|
"eval_steps_per_second": 139.565,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9810,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 7527020173056000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|