10090 lines
274 KiB
JSON
10090 lines
274 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.091696383087111,
|
|
"eval_steps": 1000,
|
|
"global_step": 5000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.832703256607056,
|
|
"epoch": 2.0417727967396844,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048573328534836735,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.1680850401520729,
|
|
"num_tokens": 4726197.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.924552583694458,
|
|
"epoch": 2.0468670402445235,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004855923727996866,
|
|
"loss": 5.5273,
|
|
"mean_token_accuracy": 0.1732455238699913,
|
|
"num_tokens": 4739238.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.825729846954346,
|
|
"epoch": 2.051961283749363,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004854507907820237,
|
|
"loss": 5.575,
|
|
"mean_token_accuracy": 0.16048659458756448,
|
|
"num_tokens": 4750658.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.91483154296875,
|
|
"epoch": 2.0570555272542026,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048530853974546495,
|
|
"loss": 5.6656,
|
|
"mean_token_accuracy": 0.16014333367347716,
|
|
"num_tokens": 4763193.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.9146116256713865,
|
|
"epoch": 2.062149770759042,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004851656201422237,
|
|
"loss": 5.6932,
|
|
"mean_token_accuracy": 0.16317821890115738,
|
|
"num_tokens": 4774701.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.883141279220581,
|
|
"epoch": 2.0672440142638817,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048502203242663846,
|
|
"loss": 5.6181,
|
|
"mean_token_accuracy": 0.1632035806775093,
|
|
"num_tokens": 4786001.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.822769355773926,
|
|
"epoch": 2.0723382577687213,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048487777705517177,
|
|
"loss": 5.5987,
|
|
"mean_token_accuracy": 0.16841727942228318,
|
|
"num_tokens": 4797591.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.937449264526367,
|
|
"epoch": 2.077432501273561,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048473285448640853,
|
|
"loss": 5.667,
|
|
"mean_token_accuracy": 0.1648013100028038,
|
|
"num_tokens": 4809296.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.849126529693604,
|
|
"epoch": 2.0825267447784004,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004845872651810549,
|
|
"loss": 5.6482,
|
|
"mean_token_accuracy": 0.16509997248649597,
|
|
"num_tokens": 4821141.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.795854616165161,
|
|
"epoch": 2.08762098828324,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048444100960193617,
|
|
"loss": 5.4898,
|
|
"mean_token_accuracy": 0.17355764359235765,
|
|
"num_tokens": 4834043.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.813658952713013,
|
|
"epoch": 2.0927152317880795,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004842940882139961,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.16684477478265763,
|
|
"num_tokens": 4845686.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.819357252120971,
|
|
"epoch": 2.097809475292919,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048414650148429465,
|
|
"loss": 5.598,
|
|
"mean_token_accuracy": 0.16547082513570785,
|
|
"num_tokens": 4857999.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.78196611404419,
|
|
"epoch": 2.1029037187977586,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048399824988200747,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.17489910870790482,
|
|
"num_tokens": 4868955.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.8591937065124515,
|
|
"epoch": 2.107997962302598,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048384933387842313,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.16161649078130721,
|
|
"num_tokens": 4881283.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.7344441413879395,
|
|
"epoch": 2.1130922058074377,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004836997539469429,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.1668286070227623,
|
|
"num_tokens": 4892841.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.882744359970093,
|
|
"epoch": 2.1181864493122773,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048354951056307814,
|
|
"loss": 5.6982,
|
|
"mean_token_accuracy": 0.16176192611455917,
|
|
"num_tokens": 4903926.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.849252843856812,
|
|
"epoch": 2.123280692817117,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048339860420444994,
|
|
"loss": 5.6035,
|
|
"mean_token_accuracy": 0.16838187575340272,
|
|
"num_tokens": 4917163.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.799989175796509,
|
|
"epoch": 2.128374936321956,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048324703535078633,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.15955802500247956,
|
|
"num_tokens": 4930232.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.853271627426148,
|
|
"epoch": 2.1334691798267955,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048309480448392185,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.17538929134607315,
|
|
"num_tokens": 4941493.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.6892719745635985,
|
|
"epoch": 2.138563423331635,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004829419120877954,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.1738343432545662,
|
|
"num_tokens": 4953781.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.899195671081543,
|
|
"epoch": 2.1436576668364746,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004827883586484488,
|
|
"loss": 5.6463,
|
|
"mean_token_accuracy": 0.16549965739250183,
|
|
"num_tokens": 4966265.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.784198760986328,
|
|
"epoch": 2.148751910341314,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004826341446540254,
|
|
"loss": 5.6537,
|
|
"mean_token_accuracy": 0.16526109278202056,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.848285675048828,
|
|
"epoch": 2.1538461538461537,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048247927059476866,
|
|
"loss": 5.6401,
|
|
"mean_token_accuracy": 0.1611713320016861,
|
|
"num_tokens": 4990818.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.839744901657104,
|
|
"epoch": 2.1589403973509933,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048232373696301985,
|
|
"loss": 5.6546,
|
|
"mean_token_accuracy": 0.16617013961076738,
|
|
"num_tokens": 5003175.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.862832260131836,
|
|
"epoch": 2.164034640855833,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048216754425321734,
|
|
"loss": 5.6158,
|
|
"mean_token_accuracy": 0.16406240314245224,
|
|
"num_tokens": 5014619.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.803761959075928,
|
|
"epoch": 2.1691288843606724,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820106929618948,
|
|
"loss": 5.6409,
|
|
"mean_token_accuracy": 0.16623104959726334,
|
|
"num_tokens": 5026407.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.875197982788086,
|
|
"epoch": 2.174223127865512,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048185318358767926,
|
|
"loss": 5.7787,
|
|
"mean_token_accuracy": 0.15349033623933792,
|
|
"num_tokens": 5039479.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.8487546920776365,
|
|
"epoch": 2.1793173713703515,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048169501663129007,
|
|
"loss": 5.6106,
|
|
"mean_token_accuracy": 0.16664023101329803,
|
|
"num_tokens": 5052074.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.991663551330566,
|
|
"epoch": 2.184411614875191,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048153619259553665,
|
|
"loss": 5.7107,
|
|
"mean_token_accuracy": 0.1539270669221878,
|
|
"num_tokens": 5064551.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.772375679016113,
|
|
"epoch": 2.1895058583800306,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048137671198531765,
|
|
"loss": 5.5901,
|
|
"mean_token_accuracy": 0.16655582785606385,
|
|
"num_tokens": 5075616.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.768369150161743,
|
|
"epoch": 2.19460010188487,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048121657530761857,
|
|
"loss": 5.6667,
|
|
"mean_token_accuracy": 0.1666068747639656,
|
|
"num_tokens": 5086531.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.914473819732666,
|
|
"epoch": 2.1996943453897098,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048105578307151094,
|
|
"loss": 5.718,
|
|
"mean_token_accuracy": 0.16477849334478378,
|
|
"num_tokens": 5097881.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.735397911071777,
|
|
"epoch": 2.2047885888945493,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048089433578814994,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.16832608878612518,
|
|
"num_tokens": 5108511.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.813470554351807,
|
|
"epoch": 2.209882832399389,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004807322339707733,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.1698839083313942,
|
|
"num_tokens": 5119302.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.78822169303894,
|
|
"epoch": 2.214977075904228,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048056947813469967,
|
|
"loss": 5.7032,
|
|
"mean_token_accuracy": 0.1617879383265972,
|
|
"num_tokens": 5131430.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.824864482879638,
|
|
"epoch": 2.220071319409068,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004804060687973264,
|
|
"loss": 5.6629,
|
|
"mean_token_accuracy": 0.16611845493316652,
|
|
"num_tokens": 5143449.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.829423427581787,
|
|
"epoch": 2.225165562913907,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004802420064781286,
|
|
"loss": 5.6386,
|
|
"mean_token_accuracy": 0.1658092647790909,
|
|
"num_tokens": 5155286.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.781202030181885,
|
|
"epoch": 2.2302598064187467,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004800772916986572,
|
|
"loss": 5.5915,
|
|
"mean_token_accuracy": 0.16370016932487488,
|
|
"num_tokens": 5167624.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.770514488220215,
|
|
"epoch": 2.235354049923586,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047991192498253715,
|
|
"loss": 5.5443,
|
|
"mean_token_accuracy": 0.17702607810497284,
|
|
"num_tokens": 5179206.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.766054010391235,
|
|
"epoch": 2.2404482934284258,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000479745906855466,
|
|
"loss": 5.6027,
|
|
"mean_token_accuracy": 0.16583019942045213,
|
|
"num_tokens": 5190820.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.834415483474731,
|
|
"epoch": 2.2455425369332653,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047957923784521206,
|
|
"loss": 5.6501,
|
|
"mean_token_accuracy": 0.16508784294128417,
|
|
"num_tokens": 5201308.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.789248037338257,
|
|
"epoch": 2.250636780438105,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004794119184816129,
|
|
"loss": 5.5475,
|
|
"mean_token_accuracy": 0.17627715319395065,
|
|
"num_tokens": 5213246.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.783004856109619,
|
|
"epoch": 2.2557310239429444,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004792439492965734,
|
|
"loss": 5.6637,
|
|
"mean_token_accuracy": 0.16591868549585342,
|
|
"num_tokens": 5225484.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.811016654968261,
|
|
"epoch": 2.260825267447784,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004790753308240642,
|
|
"loss": 5.6814,
|
|
"mean_token_accuracy": 0.1612210363149643,
|
|
"num_tokens": 5238728.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.819007301330567,
|
|
"epoch": 2.2659195109526236,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004789060636001202,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.16669994294643403,
|
|
"num_tokens": 5250130.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.835677433013916,
|
|
"epoch": 2.271013754457463,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004787361481628385,
|
|
"loss": 5.6334,
|
|
"mean_token_accuracy": 0.16798197478055954,
|
|
"num_tokens": 5262624.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.819646787643433,
|
|
"epoch": 2.2761079979623027,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004785655850523769,
|
|
"loss": 5.6789,
|
|
"mean_token_accuracy": 0.16189130991697312,
|
|
"num_tokens": 5274914.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.718485546112061,
|
|
"epoch": 2.2812022414671422,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047839437481095223,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.17636218219995498,
|
|
"num_tokens": 5286836.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.696743297576904,
|
|
"epoch": 2.286296484971982,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047822251798283836,
|
|
"loss": 5.5533,
|
|
"mean_token_accuracy": 0.16968450844287872,
|
|
"num_tokens": 5297416.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.732278060913086,
|
|
"epoch": 2.2913907284768213,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004780500151143647,
|
|
"loss": 5.5154,
|
|
"mean_token_accuracy": 0.17063294798135759,
|
|
"num_tokens": 5307852.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.76837043762207,
|
|
"epoch": 2.296484971981661,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004778768667539147,
|
|
"loss": 5.6496,
|
|
"mean_token_accuracy": 0.16280923634767533,
|
|
"num_tokens": 5320500.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.762552690505982,
|
|
"epoch": 2.3015792154865,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047770307345192326,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.16714783757925034,
|
|
"num_tokens": 5331867.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.7442131519317625,
|
|
"epoch": 2.30667345899134,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047752863576087603,
|
|
"loss": 5.5916,
|
|
"mean_token_accuracy": 0.16698855310678482,
|
|
"num_tokens": 5344181.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.864243125915527,
|
|
"epoch": 2.311767702496179,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000477353554235307,
|
|
"loss": 5.7596,
|
|
"mean_token_accuracy": 0.16341386437416078,
|
|
"num_tokens": 5356068.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.799972820281982,
|
|
"epoch": 2.3168619460010187,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047717782943179664,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.1709095746278763,
|
|
"num_tokens": 5367456.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.77769284248352,
|
|
"epoch": 2.3219561895058582,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004770014619089709,
|
|
"loss": 5.6324,
|
|
"mean_token_accuracy": 0.16938055008649827,
|
|
"num_tokens": 5379731.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.757667589187622,
|
|
"epoch": 2.327050433010698,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004768244522274985,
|
|
"loss": 5.5902,
|
|
"mean_token_accuracy": 0.16767698973417283,
|
|
"num_tokens": 5391809.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.76029486656189,
|
|
"epoch": 2.3321446765155374,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047664680095008973,
|
|
"loss": 5.6575,
|
|
"mean_token_accuracy": 0.16858063638210297,
|
|
"num_tokens": 5404341.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.790360021591186,
|
|
"epoch": 2.337238920020377,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004764685086414945,
|
|
"loss": 5.6286,
|
|
"mean_token_accuracy": 0.16352342069149017,
|
|
"num_tokens": 5415584.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.801245212554932,
|
|
"epoch": 2.3423331635252165,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047628957586850053,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.16212467104196548,
|
|
"num_tokens": 5427980.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.839998769760132,
|
|
"epoch": 2.347427407030056,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004761100031999315,
|
|
"loss": 5.6659,
|
|
"mean_token_accuracy": 0.1670482024550438,
|
|
"num_tokens": 5439782.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.727729606628418,
|
|
"epoch": 2.3525216505348956,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004759297912066454,
|
|
"loss": 5.5696,
|
|
"mean_token_accuracy": 0.16079788208007811,
|
|
"num_tokens": 5451715.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.789236354827881,
|
|
"epoch": 2.357615894039735,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047574894046153266,
|
|
"loss": 5.5259,
|
|
"mean_token_accuracy": 0.17758470326662062,
|
|
"num_tokens": 5462840.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.609377670288086,
|
|
"epoch": 2.3627101375445747,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004755674515395141,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.17512330710887908,
|
|
"num_tokens": 5474694.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.821570110321045,
|
|
"epoch": 2.3678043810494143,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004753853250175395,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.16151241809129716,
|
|
"num_tokens": 5486292.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.692107200622559,
|
|
"epoch": 2.372898624554254,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047520256147458545,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.17267432063817978,
|
|
"num_tokens": 5496799.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.794416332244873,
|
|
"epoch": 2.3779928680590934,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004750191614916536,
|
|
"loss": 5.6211,
|
|
"mean_token_accuracy": 0.17045147120952606,
|
|
"num_tokens": 5509520.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.707512474060058,
|
|
"epoch": 2.383087111563933,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004748351256517689,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.16296825110912322,
|
|
"num_tokens": 5521993.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.824226903915405,
|
|
"epoch": 2.388181355068772,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047465045453997756,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.16621532291173935,
|
|
"num_tokens": 5533878.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.760582733154297,
|
|
"epoch": 2.393275598573612,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004744651487433455,
|
|
"loss": 5.6215,
|
|
"mean_token_accuracy": 0.17181167304515838,
|
|
"num_tokens": 5546492.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.831284999847412,
|
|
"epoch": 2.398369842078451,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004742792088509561,
|
|
"loss": 5.6654,
|
|
"mean_token_accuracy": 0.16493266224861144,
|
|
"num_tokens": 5557540.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.7602368831634525,
|
|
"epoch": 2.4034640855832907,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047409263545390864,
|
|
"loss": 5.6582,
|
|
"mean_token_accuracy": 0.16328566521406174,
|
|
"num_tokens": 5569118.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.780337476730347,
|
|
"epoch": 2.4085583290881303,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004739054291453162,
|
|
"loss": 5.6831,
|
|
"mean_token_accuracy": 0.1576984331011772,
|
|
"num_tokens": 5582738.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.720583438873291,
|
|
"epoch": 2.41365257259297,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047371759052030395,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.1685176372528076,
|
|
"num_tokens": 5593941.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.750196218490601,
|
|
"epoch": 2.4187468160978094,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004735291201760072,
|
|
"loss": 5.6467,
|
|
"mean_token_accuracy": 0.15918504893779756,
|
|
"num_tokens": 5606473.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.75599160194397,
|
|
"epoch": 2.423841059602649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004733400187115694,
|
|
"loss": 5.5485,
|
|
"mean_token_accuracy": 0.17318603098392488,
|
|
"num_tokens": 5617355.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.724083518981933,
|
|
"epoch": 2.4289353031074885,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004731502867281404,
|
|
"loss": 5.6127,
|
|
"mean_token_accuracy": 0.16651729196310044,
|
|
"num_tokens": 5629062.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.839644527435302,
|
|
"epoch": 2.434029546612328,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047295992482887435,
|
|
"loss": 5.6454,
|
|
"mean_token_accuracy": 0.16396270394325257,
|
|
"num_tokens": 5641529.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.705362939834595,
|
|
"epoch": 2.4391237901171676,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047276893361892803,
|
|
"loss": 5.6807,
|
|
"mean_token_accuracy": 0.16514775156974792,
|
|
"num_tokens": 5654777.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.741829538345337,
|
|
"epoch": 2.444218033622007,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047257731370545865,
|
|
"loss": 5.6314,
|
|
"mean_token_accuracy": 0.16822979152202605,
|
|
"num_tokens": 5667208.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.667137527465821,
|
|
"epoch": 2.4493122771268467,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004723850656976221,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.17200463712215425,
|
|
"num_tokens": 5679148.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.711559009552002,
|
|
"epoch": 2.4544065206316863,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004721921902065711,
|
|
"loss": 5.4263,
|
|
"mean_token_accuracy": 0.18440213948488235,
|
|
"num_tokens": 5690869.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.656651782989502,
|
|
"epoch": 2.459500764136526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000471998687845453,
|
|
"loss": 5.6036,
|
|
"mean_token_accuracy": 0.1716683641076088,
|
|
"num_tokens": 5702939.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.819740295410156,
|
|
"epoch": 2.4645950076413654,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047180455922940783,
|
|
"loss": 5.6276,
|
|
"mean_token_accuracy": 0.16458622217178345,
|
|
"num_tokens": 5714651.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.645032453536987,
|
|
"epoch": 2.469689251146205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004716098049755668,
|
|
"loss": 5.5344,
|
|
"mean_token_accuracy": 0.16653907895088196,
|
|
"num_tokens": 5725310.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.7846427917480465,
|
|
"epoch": 2.474783494651044,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004714144257030497,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.16167016625404357,
|
|
"num_tokens": 5736357.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.731200361251831,
|
|
"epoch": 2.479877738155884,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004712184220329634,
|
|
"loss": 5.6578,
|
|
"mean_token_accuracy": 0.170462204515934,
|
|
"num_tokens": 5748281.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.8194056987762455,
|
|
"epoch": 2.484971981660723,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047102179458839965,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.1656274050474167,
|
|
"num_tokens": 5758939.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.79900860786438,
|
|
"epoch": 2.4900662251655628,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004708245439944333,
|
|
"loss": 5.6274,
|
|
"mean_token_accuracy": 0.16785999536514282,
|
|
"num_tokens": 5769750.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.753530645370484,
|
|
"epoch": 2.4951604686704023,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047062667087812007,
|
|
"loss": 5.638,
|
|
"mean_token_accuracy": 0.1693754807114601,
|
|
"num_tokens": 5781917.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.726600980758667,
|
|
"epoch": 2.500254712175242,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004704281758684946,
|
|
"loss": 5.5815,
|
|
"mean_token_accuracy": 0.16392055451869963,
|
|
"num_tokens": 5793665.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.7634521484375,
|
|
"epoch": 2.5053489556800814,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004702290595965686,
|
|
"loss": 5.6906,
|
|
"mean_token_accuracy": 0.1637534260749817,
|
|
"num_tokens": 5806364.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.799836015701294,
|
|
"epoch": 2.510443199184921,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004700293226953288,
|
|
"loss": 5.678,
|
|
"mean_token_accuracy": 0.16520603075623513,
|
|
"num_tokens": 5818693.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.598745918273925,
|
|
"epoch": 2.5155374426897605,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004698289657997348,
|
|
"loss": 5.4413,
|
|
"mean_token_accuracy": 0.18404357582330705,
|
|
"num_tokens": 5828765.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.733962345123291,
|
|
"epoch": 2.5206316861946,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046962798954671727,
|
|
"loss": 5.6088,
|
|
"mean_token_accuracy": 0.174197518825531,
|
|
"num_tokens": 5840196.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.758548879623413,
|
|
"epoch": 2.5257259296994397,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046942639457517565,
|
|
"loss": 5.5993,
|
|
"mean_token_accuracy": 0.1715373769402504,
|
|
"num_tokens": 5851117.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.70658049583435,
|
|
"epoch": 2.530820173204279,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004692241815259765,
|
|
"loss": 5.5836,
|
|
"mean_token_accuracy": 0.16994259804487227,
|
|
"num_tokens": 5862918.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.7033356666564945,
|
|
"epoch": 2.5359144167091188,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000469021351041951,
|
|
"loss": 5.6134,
|
|
"mean_token_accuracy": 0.16869816184043884,
|
|
"num_tokens": 5875620.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.795348024368286,
|
|
"epoch": 2.5410086602139583,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004688179037678933,
|
|
"loss": 5.6272,
|
|
"mean_token_accuracy": 0.16622024178504943,
|
|
"num_tokens": 5887113.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.740134620666504,
|
|
"epoch": 2.546102903718798,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004686138403505582,
|
|
"loss": 5.6687,
|
|
"mean_token_accuracy": 0.16488093435764312,
|
|
"num_tokens": 5898707.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.797619247436524,
|
|
"epoch": 2.5511971472236374,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046840916143865913,
|
|
"loss": 5.5948,
|
|
"mean_token_accuracy": 0.1700372129678726,
|
|
"num_tokens": 5909548.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.742323207855224,
|
|
"epoch": 2.556291390728477,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004682038676828666,
|
|
"loss": 5.5575,
|
|
"mean_token_accuracy": 0.17342328727245332,
|
|
"num_tokens": 5920750.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.703403377532959,
|
|
"epoch": 2.561385634233316,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046799795973580515,
|
|
"loss": 5.6814,
|
|
"mean_token_accuracy": 0.16718617379665374,
|
|
"num_tokens": 5932832.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.757970190048217,
|
|
"epoch": 2.566479877738156,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046779143825205217,
|
|
"loss": 5.6053,
|
|
"mean_token_accuracy": 0.17122609466314315,
|
|
"num_tokens": 5945644.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.6888148307800295,
|
|
"epoch": 2.5715741212429952,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004675843038881354,
|
|
"loss": 5.6015,
|
|
"mean_token_accuracy": 0.16385104954242707,
|
|
"num_tokens": 5956760.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.801866722106934,
|
|
"epoch": 2.576668364747835,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046737655730253076,
|
|
"loss": 5.6063,
|
|
"mean_token_accuracy": 0.17185162752866745,
|
|
"num_tokens": 5969179.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.7418070316314695,
|
|
"epoch": 2.5817626082526743,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046716819915566067,
|
|
"loss": 5.6916,
|
|
"mean_token_accuracy": 0.16401710957288743,
|
|
"num_tokens": 5982051.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.8431535243988035,
|
|
"epoch": 2.586856851757514,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046695923010989145,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.15622967928647996,
|
|
"num_tokens": 5994038.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.820972156524658,
|
|
"epoch": 2.5919510952623535,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046674965082953157,
|
|
"loss": 5.6269,
|
|
"mean_token_accuracy": 0.16563534885644912,
|
|
"num_tokens": 6005356.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.726960706710815,
|
|
"epoch": 2.597045338767193,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046653946198082943,
|
|
"loss": 5.6612,
|
|
"mean_token_accuracy": 0.16744568645954133,
|
|
"num_tokens": 6018477.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.697501373291016,
|
|
"epoch": 2.6021395822720326,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046632866423197125,
|
|
"loss": 5.5698,
|
|
"mean_token_accuracy": 0.1735507518053055,
|
|
"num_tokens": 6029084.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.74003415107727,
|
|
"epoch": 2.607233825776872,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004661172582530788,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.1741526886820793,
|
|
"num_tokens": 6040271.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.630206871032715,
|
|
"epoch": 2.6123280692817117,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004659052447162076,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.18263173550367356,
|
|
"num_tokens": 6051278.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.6890685081481935,
|
|
"epoch": 2.6174223127865512,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004656926242953445,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.16885432302951814,
|
|
"num_tokens": 6063397.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.774868965148926,
|
|
"epoch": 2.622516556291391,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004654793976664055,
|
|
"loss": 5.5693,
|
|
"mean_token_accuracy": 0.17261110693216325,
|
|
"num_tokens": 6075105.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.774467945098877,
|
|
"epoch": 2.6276107997962304,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046526556550723395,
|
|
"loss": 5.6196,
|
|
"mean_token_accuracy": 0.16647132039070128,
|
|
"num_tokens": 6086069.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.68480806350708,
|
|
"epoch": 2.63270504330107,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046505112849759794,
|
|
"loss": 5.5567,
|
|
"mean_token_accuracy": 0.1698553279042244,
|
|
"num_tokens": 6096997.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.6718714237213135,
|
|
"epoch": 2.6377992868059095,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046483608731918855,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.17409039437770843,
|
|
"num_tokens": 6109510.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.712036037445069,
|
|
"epoch": 2.642893530310749,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046462044265561735,
|
|
"loss": 5.4953,
|
|
"mean_token_accuracy": 0.17595480382442474,
|
|
"num_tokens": 6120111.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.718178939819336,
|
|
"epoch": 2.647987773815588,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046440419519241455,
|
|
"loss": 5.653,
|
|
"mean_token_accuracy": 0.16612613946199417,
|
|
"num_tokens": 6131558.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.776505422592163,
|
|
"epoch": 2.653082017320428,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004641873456170265,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.17298970967531205,
|
|
"num_tokens": 6143687.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.804161024093628,
|
|
"epoch": 2.6581762608252673,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004639698946188137,
|
|
"loss": 5.6962,
|
|
"mean_token_accuracy": 0.1682490512728691,
|
|
"num_tokens": 6153692.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.703200149536133,
|
|
"epoch": 2.663270504330107,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004637518428890486,
|
|
"loss": 5.5836,
|
|
"mean_token_accuracy": 0.17761685699224472,
|
|
"num_tokens": 6164689.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.752318668365478,
|
|
"epoch": 2.6683647478349464,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004635331911209133,
|
|
"loss": 5.6502,
|
|
"mean_token_accuracy": 0.1659931257367134,
|
|
"num_tokens": 6176720.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.705931329727173,
|
|
"epoch": 2.673458991339786,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004633139400094974,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.17467646300792694,
|
|
"num_tokens": 6187745.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.672006559371948,
|
|
"epoch": 2.6785532348446255,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000463094090251796,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.17560752034187316,
|
|
"num_tokens": 6200403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.671006202697754,
|
|
"epoch": 2.683647478349465,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046287364254670696,
|
|
"loss": 5.559,
|
|
"mean_token_accuracy": 0.17330682277679443,
|
|
"num_tokens": 6211412.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.8153105735778805,
|
|
"epoch": 2.6887417218543046,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004626525975950292,
|
|
"loss": 5.6857,
|
|
"mean_token_accuracy": 0.1614910453557968,
|
|
"num_tokens": 6222798.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.695845890045166,
|
|
"epoch": 2.693835965359144,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004624309560994603,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.1753826379776001,
|
|
"num_tokens": 6234846.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.689202404022216,
|
|
"epoch": 2.6989302088639837,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004622087187645942,
|
|
"loss": 5.5786,
|
|
"mean_token_accuracy": 0.17166780084371566,
|
|
"num_tokens": 6245771.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.675031471252441,
|
|
"epoch": 2.7040244523688233,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004619858862969188,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.17476927042007445,
|
|
"num_tokens": 6258254.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.722859048843384,
|
|
"epoch": 2.709118695873663,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004617624594048143,
|
|
"loss": 5.5905,
|
|
"mean_token_accuracy": 0.1723139241337776,
|
|
"num_tokens": 6269383.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.832469272613525,
|
|
"epoch": 2.7142129393785024,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004615384387985504,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.17347706407308577,
|
|
"num_tokens": 6280275.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.597794389724731,
|
|
"epoch": 2.719307182883342,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000461313825190284,
|
|
"loss": 5.5715,
|
|
"mean_token_accuracy": 0.164817114174366,
|
|
"num_tokens": 6291760.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.6889818668365475,
|
|
"epoch": 2.7244014263881815,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004610886192940574,
|
|
"loss": 5.5343,
|
|
"mean_token_accuracy": 0.17360015511512755,
|
|
"num_tokens": 6303041.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.69299807548523,
|
|
"epoch": 2.729495669893021,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046086282182579566,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.1723339691758156,
|
|
"num_tokens": 6313950.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.559491395950317,
|
|
"epoch": 2.73458991339786,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004606364335033045,
|
|
"loss": 5.6334,
|
|
"mean_token_accuracy": 0.16981981843709945,
|
|
"num_tokens": 6325830.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.856181812286377,
|
|
"epoch": 2.7396841569027,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046040945504626787,
|
|
"loss": 5.6731,
|
|
"mean_token_accuracy": 0.16822368204593657,
|
|
"num_tokens": 6337978.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.664763259887695,
|
|
"epoch": 2.7447784004075393,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046018188717624576,
|
|
"loss": 5.5185,
|
|
"mean_token_accuracy": 0.17573055624961853,
|
|
"num_tokens": 6349623.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.628146028518676,
|
|
"epoch": 2.7498726439123793,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000459953730616672,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.17330321222543715,
|
|
"num_tokens": 6361149.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.65805139541626,
|
|
"epoch": 2.7549668874172184,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045972498609285175,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.17333125770092012,
|
|
"num_tokens": 6372691.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.715612506866455,
|
|
"epoch": 2.760061130922058,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045949565433195925,
|
|
"loss": 5.6554,
|
|
"mean_token_accuracy": 0.16047533005475997,
|
|
"num_tokens": 6385560.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.728322315216064,
|
|
"epoch": 2.7651553744268975,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045926573606303576,
|
|
"loss": 5.5959,
|
|
"mean_token_accuracy": 0.16160788387060165,
|
|
"num_tokens": 6397227.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.748125410079956,
|
|
"epoch": 2.770249617931737,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045903523201698684,
|
|
"loss": 5.6109,
|
|
"mean_token_accuracy": 0.1713184103369713,
|
|
"num_tokens": 6409906.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.760515928268433,
|
|
"epoch": 2.7753438614365766,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004588041429265803,
|
|
"loss": 5.5729,
|
|
"mean_token_accuracy": 0.1721328616142273,
|
|
"num_tokens": 6421922.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.710689926147461,
|
|
"epoch": 2.780438104941416,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004585724695264436,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.16782642304897308,
|
|
"num_tokens": 6433135.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.758449697494507,
|
|
"epoch": 2.7855323484462557,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045834021255306223,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.16317217350006102,
|
|
"num_tokens": 6444399.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.7549810886383055,
|
|
"epoch": 2.7906265919510953,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004581073727447762,
|
|
"loss": 5.6676,
|
|
"mean_token_accuracy": 0.16437004804611205,
|
|
"num_tokens": 6456227.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.690370273590088,
|
|
"epoch": 2.795720835455935,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004578739508417788,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.17375921159982682,
|
|
"num_tokens": 6468622.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.688498878479004,
|
|
"epoch": 2.8008150789607744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004576399475861137,
|
|
"loss": 5.609,
|
|
"mean_token_accuracy": 0.1656633049249649,
|
|
"num_tokens": 6480408.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.72419810295105,
|
|
"epoch": 2.805909322465614,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004574053637216724,
|
|
"loss": 5.5493,
|
|
"mean_token_accuracy": 0.1693354807794094,
|
|
"num_tokens": 6492864.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.614224910736084,
|
|
"epoch": 2.8110035659704535,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004571701999941926,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.1756269544363022,
|
|
"num_tokens": 6504103.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.742954730987549,
|
|
"epoch": 2.816097809475293,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000456934457151255,
|
|
"loss": 5.6366,
|
|
"mean_token_accuracy": 0.16199332922697068,
|
|
"num_tokens": 6516560.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.617284107208252,
|
|
"epoch": 2.821192052980132,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045669813594228137,
|
|
"loss": 5.5866,
|
|
"mean_token_accuracy": 0.17059138864278794,
|
|
"num_tokens": 6530002.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.758786964416504,
|
|
"epoch": 2.826286296484972,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045646123711853235,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.16911731660366058,
|
|
"num_tokens": 6542454.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.5954100608825685,
|
|
"epoch": 2.8313805399898113,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045622376143310435,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.17661769539117814,
|
|
"num_tokens": 6553385.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.69018235206604,
|
|
"epoch": 2.8364747834946513,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045598570964092817,
|
|
"loss": 5.5222,
|
|
"mean_token_accuracy": 0.17083494812250138,
|
|
"num_tokens": 6564385.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.669094085693359,
|
|
"epoch": 2.8415690269994904,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045574708249876547,
|
|
"loss": 5.514,
|
|
"mean_token_accuracy": 0.17221893221139908,
|
|
"num_tokens": 6575369.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.6958842277526855,
|
|
"epoch": 2.84666327050433,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045550788076520736,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.17091407477855683,
|
|
"num_tokens": 6587173.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.795775032043457,
|
|
"epoch": 2.8517575140091695,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004552681052006713,
|
|
"loss": 5.6766,
|
|
"mean_token_accuracy": 0.16159882098436357,
|
|
"num_tokens": 6598919.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.729952669143676,
|
|
"epoch": 2.856851757514009,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004550277565673992,
|
|
"loss": 5.6106,
|
|
"mean_token_accuracy": 0.17154622673988343,
|
|
"num_tokens": 6610353.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.618022871017456,
|
|
"epoch": 2.8619460010188487,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045478683562945457,
|
|
"loss": 5.5707,
|
|
"mean_token_accuracy": 0.1710265427827835,
|
|
"num_tokens": 6622420.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.724002742767334,
|
|
"epoch": 2.867040244523688,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004545453431527202,
|
|
"loss": 5.5908,
|
|
"mean_token_accuracy": 0.16568071395158768,
|
|
"num_tokens": 6634234.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.818776416778564,
|
|
"epoch": 2.8721344880285278,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000454303279904896,
|
|
"loss": 5.6997,
|
|
"mean_token_accuracy": 0.16635357588529587,
|
|
"num_tokens": 6648272.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.690582895278931,
|
|
"epoch": 2.8772287315333673,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045406064665549624,
|
|
"loss": 5.5177,
|
|
"mean_token_accuracy": 0.1796009883284569,
|
|
"num_tokens": 6660114.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.687472009658814,
|
|
"epoch": 2.882322975038207,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004538174441758471,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.16835884898900985,
|
|
"num_tokens": 6670433.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.590880060195923,
|
|
"epoch": 2.8874172185430464,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045357367323908463,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.17777381986379623,
|
|
"num_tokens": 6681918.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.8125121116638185,
|
|
"epoch": 2.892511462047886,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045332933462015156,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.16489193439483643,
|
|
"num_tokens": 6692681.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.686181831359863,
|
|
"epoch": 2.8976057055527256,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004530844290957957,
|
|
"loss": 5.5918,
|
|
"mean_token_accuracy": 0.16602931022644044,
|
|
"num_tokens": 6704454.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.687756729125977,
|
|
"epoch": 2.902699949057565,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045283895744456694,
|
|
"loss": 5.563,
|
|
"mean_token_accuracy": 0.16700583547353745,
|
|
"num_tokens": 6718226.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.6581353664398195,
|
|
"epoch": 2.9077941925624042,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045259292044681456,
|
|
"loss": 5.5292,
|
|
"mean_token_accuracy": 0.17562829554080964,
|
|
"num_tokens": 6730507.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.646629428863525,
|
|
"epoch": 2.9128884360672442,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004523463188846855,
|
|
"loss": 5.5816,
|
|
"mean_token_accuracy": 0.16764041483402253,
|
|
"num_tokens": 6743312.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.6718567371368405,
|
|
"epoch": 2.9179826795720833,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004520991535421211,
|
|
"loss": 5.5671,
|
|
"mean_token_accuracy": 0.17292603999376296,
|
|
"num_tokens": 6757945.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.6956487655639645,
|
|
"epoch": 2.9230769230769234,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004518514252048552,
|
|
"loss": 5.6138,
|
|
"mean_token_accuracy": 0.16357647478580475,
|
|
"num_tokens": 6770995.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.695530366897583,
|
|
"epoch": 2.9281711665817625,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004516031346604112,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.17252999693155288,
|
|
"num_tokens": 6782117.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.60226001739502,
|
|
"epoch": 2.933265410086602,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004513542826980999,
|
|
"loss": 5.6264,
|
|
"mean_token_accuracy": 0.1690582275390625,
|
|
"num_tokens": 6795932.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.7579139232635494,
|
|
"epoch": 2.9383596535914416,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004511048701090167,
|
|
"loss": 5.531,
|
|
"mean_token_accuracy": 0.17775799632072448,
|
|
"num_tokens": 6805708.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.73887186050415,
|
|
"epoch": 2.943453897096281,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004508548976860392,
|
|
"loss": 5.5859,
|
|
"mean_token_accuracy": 0.16735829114913942,
|
|
"num_tokens": 6816964.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.726815509796142,
|
|
"epoch": 2.9485481406011207,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000450604366223825,
|
|
"loss": 5.6149,
|
|
"mean_token_accuracy": 0.16673005521297454,
|
|
"num_tokens": 6828330.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.723990058898925,
|
|
"epoch": 2.9536423841059603,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045035327651880846,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.16775917112827302,
|
|
"num_tokens": 6839860.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.686678028106689,
|
|
"epoch": 2.9587366276108,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045010162936919896,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.17354528307914735,
|
|
"num_tokens": 6850453.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.635978555679321,
|
|
"epoch": 2.9638308711156394,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004498494255749776,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.17512332648038864,
|
|
"num_tokens": 6861661.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.708203220367432,
|
|
"epoch": 2.968925114620479,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004495966659378954,
|
|
"loss": 5.5989,
|
|
"mean_token_accuracy": 0.17088168114423752,
|
|
"num_tokens": 6873474.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.683584547042846,
|
|
"epoch": 2.9740193581253185,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044934335126147016,
|
|
"loss": 5.545,
|
|
"mean_token_accuracy": 0.17199764996767045,
|
|
"num_tokens": 6885761.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.67879090309143,
|
|
"epoch": 2.979113601630158,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004490894823509843,
|
|
"loss": 5.5667,
|
|
"mean_token_accuracy": 0.16536117047071458,
|
|
"num_tokens": 6895911.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.707906007766724,
|
|
"epoch": 2.9842078451349976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004488350600134819,
|
|
"loss": 5.5051,
|
|
"mean_token_accuracy": 0.1784601166844368,
|
|
"num_tokens": 6907431.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.593939113616943,
|
|
"epoch": 2.989302088639837,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004485800850577667,
|
|
"loss": 5.5797,
|
|
"mean_token_accuracy": 0.17540695071220397,
|
|
"num_tokens": 6919007.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.7369384765625,
|
|
"epoch": 2.9943963321446763,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004483245582943989,
|
|
"loss": 5.5901,
|
|
"mean_token_accuracy": 0.17304329574108124,
|
|
"num_tokens": 6930194.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.652069902420044,
|
|
"epoch": 2.9994905756495163,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044806848053569314,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.17087605148553847,
|
|
"num_tokens": 6941276.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.668265342712402,
|
|
"epoch": 3.0040753948038716,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044781185259571544,
|
|
"loss": 5.2826,
|
|
"mean_token_accuracy": 0.19153548777103424,
|
|
"num_tokens": 6952351.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.731490230560302,
|
|
"epoch": 3.009169638308711,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004475546752902811,
|
|
"loss": 5.296,
|
|
"mean_token_accuracy": 0.18416041433811187,
|
|
"num_tokens": 6965052.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.547205829620362,
|
|
"epoch": 3.0142638818135508,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044729694943695155,
|
|
"loss": 5.164,
|
|
"mean_token_accuracy": 0.19336679726839065,
|
|
"num_tokens": 6976662.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.639112377166748,
|
|
"epoch": 3.0193581253183903,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004470386758550321,
|
|
"loss": 5.2328,
|
|
"mean_token_accuracy": 0.18733217120170592,
|
|
"num_tokens": 6989787.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.647245597839356,
|
|
"epoch": 3.02445236882323,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004467798553655695,
|
|
"loss": 5.1776,
|
|
"mean_token_accuracy": 0.18518314361572266,
|
|
"num_tokens": 7000866.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.641472673416137,
|
|
"epoch": 3.0295466123280694,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000446520488791349,
|
|
"loss": 5.2815,
|
|
"mean_token_accuracy": 0.18436826020479202,
|
|
"num_tokens": 7012862.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.654988145828247,
|
|
"epoch": 3.034640855832909,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004462605769568915,
|
|
"loss": 5.2541,
|
|
"mean_token_accuracy": 0.1877979502081871,
|
|
"num_tokens": 7025434.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.604527473449707,
|
|
"epoch": 3.0397350993377485,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004460001206884519,
|
|
"loss": 5.2583,
|
|
"mean_token_accuracy": 0.18266716599464417,
|
|
"num_tokens": 7037458.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.622461557388306,
|
|
"epoch": 3.0448293428425877,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004457391208140155,
|
|
"loss": 5.2308,
|
|
"mean_token_accuracy": 0.19294089525938035,
|
|
"num_tokens": 7049493.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.606506872177124,
|
|
"epoch": 3.049923586347427,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004454775781632955,
|
|
"loss": 5.2567,
|
|
"mean_token_accuracy": 0.18151833415031432,
|
|
"num_tokens": 7059824.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.595340728759766,
|
|
"epoch": 3.0550178298522668,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044521549356773096,
|
|
"loss": 5.179,
|
|
"mean_token_accuracy": 0.1940339609980583,
|
|
"num_tokens": 7070604.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0550178298522668,
|
|
"eval_entropy": 5.490248529410657,
|
|
"eval_loss": 5.822233200073242,
|
|
"eval_mean_token_accuracy": 0.16649478103643583,
|
|
"eval_num_tokens": 7070604.0,
|
|
"eval_runtime": 2.6541,
|
|
"eval_samples_per_second": 1220.744,
|
|
"eval_steps_per_second": 152.593,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.5521571159362795,
|
|
"epoch": 3.0601120733571063,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004449528678604836,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.18909391462802888,
|
|
"num_tokens": 7083166.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.669453763961792,
|
|
"epoch": 3.065206316861946,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004446897018764353,
|
|
"loss": 5.2677,
|
|
"mean_token_accuracy": 0.1855538919568062,
|
|
"num_tokens": 7094749.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.5387187004089355,
|
|
"epoch": 3.0703005603667854,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044442599645218566,
|
|
"loss": 5.2142,
|
|
"mean_token_accuracy": 0.18410991579294206,
|
|
"num_tokens": 7105635.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.593394804000854,
|
|
"epoch": 3.075394803871625,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044416175242604887,
|
|
"loss": 5.261,
|
|
"mean_token_accuracy": 0.18290836960077286,
|
|
"num_tokens": 7116972.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.50187931060791,
|
|
"epoch": 3.0804890473764646,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004438969706380514,
|
|
"loss": 5.1523,
|
|
"mean_token_accuracy": 0.19758753180503846,
|
|
"num_tokens": 7128085.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.578138923645019,
|
|
"epoch": 3.085583290881304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044363165192992947,
|
|
"loss": 5.2348,
|
|
"mean_token_accuracy": 0.18552803695201875,
|
|
"num_tokens": 7139587.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.698646783828735,
|
|
"epoch": 3.0906775343861437,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004433657971451259,
|
|
"loss": 5.3839,
|
|
"mean_token_accuracy": 0.173110431432724,
|
|
"num_tokens": 7152227.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.571455335617065,
|
|
"epoch": 3.0957717778909832,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044309940712878774,
|
|
"loss": 5.2578,
|
|
"mean_token_accuracy": 0.18726434856653212,
|
|
"num_tokens": 7164203.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.602433776855468,
|
|
"epoch": 3.100866021395823,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004428324827277637,
|
|
"loss": 5.2718,
|
|
"mean_token_accuracy": 0.19115810692310334,
|
|
"num_tokens": 7175678.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.52528223991394,
|
|
"epoch": 3.1059602649006623,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000442565024790601,
|
|
"loss": 5.18,
|
|
"mean_token_accuracy": 0.19296997636556626,
|
|
"num_tokens": 7187482.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.584066247940063,
|
|
"epoch": 3.111054508405502,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044229703416754333,
|
|
"loss": 5.2521,
|
|
"mean_token_accuracy": 0.18557969182729722,
|
|
"num_tokens": 7199198.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.602118587493896,
|
|
"epoch": 3.1161487519103415,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004420285117105275,
|
|
"loss": 5.229,
|
|
"mean_token_accuracy": 0.19193842113018036,
|
|
"num_tokens": 7209877.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.547281932830811,
|
|
"epoch": 3.121242995415181,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044175945827318097,
|
|
"loss": 5.238,
|
|
"mean_token_accuracy": 0.18117648512125015,
|
|
"num_tokens": 7221838.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.5285964012146,
|
|
"epoch": 3.1263372389200206,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044148987471081933,
|
|
"loss": 5.2705,
|
|
"mean_token_accuracy": 0.18398284912109375,
|
|
"num_tokens": 7233675.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.557986259460449,
|
|
"epoch": 3.13143148242486,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044121976188044357,
|
|
"loss": 5.1749,
|
|
"mean_token_accuracy": 0.20045067518949508,
|
|
"num_tokens": 7243888.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.505409860610962,
|
|
"epoch": 3.1365257259296992,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004409491206407367,
|
|
"loss": 5.2605,
|
|
"mean_token_accuracy": 0.18896156549453735,
|
|
"num_tokens": 7256157.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.592403697967529,
|
|
"epoch": 3.141619969434539,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004406779518520622,
|
|
"loss": 5.2726,
|
|
"mean_token_accuracy": 0.18226876109838486,
|
|
"num_tokens": 7268270.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.634896612167358,
|
|
"epoch": 3.1467142129393784,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044040625637646006,
|
|
"loss": 5.2992,
|
|
"mean_token_accuracy": 0.1847335711121559,
|
|
"num_tokens": 7281031.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.517871999740601,
|
|
"epoch": 3.151808456444218,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004401340350776449,
|
|
"loss": 5.171,
|
|
"mean_token_accuracy": 0.1987476795911789,
|
|
"num_tokens": 7291891.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.548020076751709,
|
|
"epoch": 3.1569026999490575,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043986128882100286,
|
|
"loss": 5.2714,
|
|
"mean_token_accuracy": 0.18431186974048613,
|
|
"num_tokens": 7303329.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.589869499206543,
|
|
"epoch": 3.161996943453897,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043958801847358874,
|
|
"loss": 5.2381,
|
|
"mean_token_accuracy": 0.19727718085050583,
|
|
"num_tokens": 7315517.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.530579137802124,
|
|
"epoch": 3.1670911869587366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043931422490412376,
|
|
"loss": 5.1944,
|
|
"mean_token_accuracy": 0.1873103693127632,
|
|
"num_tokens": 7325795.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.562802267074585,
|
|
"epoch": 3.172185430463576,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043903990898299213,
|
|
"loss": 5.169,
|
|
"mean_token_accuracy": 0.1867876723408699,
|
|
"num_tokens": 7337587.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.5110859870910645,
|
|
"epoch": 3.1772796739684157,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043876507158223865,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.1782894030213356,
|
|
"num_tokens": 7349806.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.5876891136169435,
|
|
"epoch": 3.1823739174732553,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004384897135755663,
|
|
"loss": 5.2945,
|
|
"mean_token_accuracy": 0.17804135233163834,
|
|
"num_tokens": 7362544.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.49224066734314,
|
|
"epoch": 3.187468160978095,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043821383583833237,
|
|
"loss": 5.1725,
|
|
"mean_token_accuracy": 0.19039895087480546,
|
|
"num_tokens": 7374931.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.5407790184021,
|
|
"epoch": 3.1925624044829344,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004379374392475468,
|
|
"loss": 5.2126,
|
|
"mean_token_accuracy": 0.18609565496444702,
|
|
"num_tokens": 7385609.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.551992416381836,
|
|
"epoch": 3.197656647987774,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004376605246818691,
|
|
"loss": 5.2421,
|
|
"mean_token_accuracy": 0.1855844110250473,
|
|
"num_tokens": 7396492.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.492816591262818,
|
|
"epoch": 3.2027508914926135,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437383093021605,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.19590026140213013,
|
|
"num_tokens": 7409319.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.548036813735962,
|
|
"epoch": 3.207845134997453,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004371051451487042,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.185629640519619,
|
|
"num_tokens": 7420395.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.567162942886353,
|
|
"epoch": 3.2129393785022926,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043682668194675766,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.17593613266944885,
|
|
"num_tokens": 7432949.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.516055631637573,
|
|
"epoch": 3.218033622007132,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004365477043009943,
|
|
"loss": 5.221,
|
|
"mean_token_accuracy": 0.1935055822134018,
|
|
"num_tokens": 7444849.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.520792007446289,
|
|
"epoch": 3.2231278655119713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043626821309827846,
|
|
"loss": 5.2801,
|
|
"mean_token_accuracy": 0.1806574985384941,
|
|
"num_tokens": 7455736.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.535524129867554,
|
|
"epoch": 3.228222109016811,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043598820922710727,
|
|
"loss": 5.1924,
|
|
"mean_token_accuracy": 0.1915664106607437,
|
|
"num_tokens": 7466711.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.575295257568359,
|
|
"epoch": 3.2333163525216504,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043570769357760737,
|
|
"loss": 5.2477,
|
|
"mean_token_accuracy": 0.18594499677419662,
|
|
"num_tokens": 7478159.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.507748985290528,
|
|
"epoch": 3.23841059602649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043542666704153257,
|
|
"loss": 5.2252,
|
|
"mean_token_accuracy": 0.1880890280008316,
|
|
"num_tokens": 7488657.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.591403770446777,
|
|
"epoch": 3.2435048395313295,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004351451305122605,
|
|
"loss": 5.3204,
|
|
"mean_token_accuracy": 0.1818707376718521,
|
|
"num_tokens": 7499949.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.485433578491211,
|
|
"epoch": 3.248599083036169,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043486308488479045,
|
|
"loss": 5.1313,
|
|
"mean_token_accuracy": 0.1981205612421036,
|
|
"num_tokens": 7511013.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.530738973617554,
|
|
"epoch": 3.2536933265410086,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004345805310557397,
|
|
"loss": 5.2657,
|
|
"mean_token_accuracy": 0.18974434435367585,
|
|
"num_tokens": 7522325.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.5567710399627686,
|
|
"epoch": 3.258787570045848,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004342974699233412,
|
|
"loss": 5.2639,
|
|
"mean_token_accuracy": 0.18874050825834274,
|
|
"num_tokens": 7534249.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.63561372756958,
|
|
"epoch": 3.2638818135506877,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043401390238744094,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.1869332253932953,
|
|
"num_tokens": 7545472.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.645912122726441,
|
|
"epoch": 3.2689760570555273,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004337298293494943,
|
|
"loss": 5.35,
|
|
"mean_token_accuracy": 0.18099952787160872,
|
|
"num_tokens": 7557188.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.497475528717041,
|
|
"epoch": 3.274070300560367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043344525171256395,
|
|
"loss": 5.2865,
|
|
"mean_token_accuracy": 0.1850983291864395,
|
|
"num_tokens": 7569464.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.615909624099731,
|
|
"epoch": 3.2791645440652064,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004331601703813165,
|
|
"loss": 5.2998,
|
|
"mean_token_accuracy": 0.1851735457777977,
|
|
"num_tokens": 7581314.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.591563701629639,
|
|
"epoch": 3.284258787570046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004328745862620198,
|
|
"loss": 5.3103,
|
|
"mean_token_accuracy": 0.1866564929485321,
|
|
"num_tokens": 7593419.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.518194627761841,
|
|
"epoch": 3.2893530310748855,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004325885002625403,
|
|
"loss": 5.2704,
|
|
"mean_token_accuracy": 0.18611360490322112,
|
|
"num_tokens": 7604952.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.517184400558472,
|
|
"epoch": 3.294447274579725,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004323019132923396,
|
|
"loss": 5.2068,
|
|
"mean_token_accuracy": 0.19184933006763458,
|
|
"num_tokens": 7616634.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.53750467300415,
|
|
"epoch": 3.299541518084564,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004320148262624719,
|
|
"loss": 5.3194,
|
|
"mean_token_accuracy": 0.18067368865013123,
|
|
"num_tokens": 7628628.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.564935636520386,
|
|
"epoch": 3.304635761589404,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004317272400855814,
|
|
"loss": 5.2302,
|
|
"mean_token_accuracy": 0.19101853519678116,
|
|
"num_tokens": 7639211.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.498926210403442,
|
|
"epoch": 3.3097300050942433,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043143915567589875,
|
|
"loss": 5.2103,
|
|
"mean_token_accuracy": 0.19033780992031096,
|
|
"num_tokens": 7649985.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.525574207305908,
|
|
"epoch": 3.314824248599083,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004311505739492385,
|
|
"loss": 5.3175,
|
|
"mean_token_accuracy": 0.18325382769107817,
|
|
"num_tokens": 7661280.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.5513904094696045,
|
|
"epoch": 3.3199184921039224,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043086149582299637,
|
|
"loss": 5.2999,
|
|
"mean_token_accuracy": 0.18504323214292526,
|
|
"num_tokens": 7674588.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.5241905689239506,
|
|
"epoch": 3.325012735608762,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043057192221614604,
|
|
"loss": 5.2586,
|
|
"mean_token_accuracy": 0.19100991785526275,
|
|
"num_tokens": 7687481.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.567141199111939,
|
|
"epoch": 3.3301069791136015,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043028185404923614,
|
|
"loss": 5.3418,
|
|
"mean_token_accuracy": 0.18440684229135512,
|
|
"num_tokens": 7699536.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.503921937942505,
|
|
"epoch": 3.335201222618441,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004299912922443877,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.1870080202817917,
|
|
"num_tokens": 7711135.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.547470760345459,
|
|
"epoch": 3.3402954661232807,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004297002377252908,
|
|
"loss": 5.2792,
|
|
"mean_token_accuracy": 0.19307379573583602,
|
|
"num_tokens": 7722385.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.474522304534912,
|
|
"epoch": 3.34538970962812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042940869141720224,
|
|
"loss": 5.2535,
|
|
"mean_token_accuracy": 0.18212098628282547,
|
|
"num_tokens": 7735177.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.544339656829834,
|
|
"epoch": 3.3504839531329598,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004291166542469418,
|
|
"loss": 5.2564,
|
|
"mean_token_accuracy": 0.18411953747272491,
|
|
"num_tokens": 7747245.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.622571277618408,
|
|
"epoch": 3.3555781966377993,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042882412714288987,
|
|
"loss": 5.3085,
|
|
"mean_token_accuracy": 0.18051094710826873,
|
|
"num_tokens": 7759267.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.529182434082031,
|
|
"epoch": 3.360672440142639,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042853111103498433,
|
|
"loss": 5.2604,
|
|
"mean_token_accuracy": 0.18174230456352233,
|
|
"num_tokens": 7769973.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.581902551651001,
|
|
"epoch": 3.3657666836474784,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042823760685471756,
|
|
"loss": 5.2093,
|
|
"mean_token_accuracy": 0.19707539081573486,
|
|
"num_tokens": 7780249.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.536277008056641,
|
|
"epoch": 3.370860927152318,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004279436155351336,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.19473234415054322,
|
|
"num_tokens": 7792240.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.5104621887207035,
|
|
"epoch": 3.3759551706571576,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004276491380108251,
|
|
"loss": 5.2588,
|
|
"mean_token_accuracy": 0.18793881684541702,
|
|
"num_tokens": 7804996.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.570809698104858,
|
|
"epoch": 3.381049414161997,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004273541752179301,
|
|
"loss": 5.2978,
|
|
"mean_token_accuracy": 0.18294053822755812,
|
|
"num_tokens": 7816980.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.529162979125976,
|
|
"epoch": 3.3861436576668362,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042705872809412955,
|
|
"loss": 5.3187,
|
|
"mean_token_accuracy": 0.18205417394638063,
|
|
"num_tokens": 7829401.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.613763236999512,
|
|
"epoch": 3.3912379011716762,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042676279757864404,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.1894928053021431,
|
|
"num_tokens": 7843526.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.461062240600586,
|
|
"epoch": 3.3963321446765153,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042646638461223076,
|
|
"loss": 5.146,
|
|
"mean_token_accuracy": 0.20297018587589263,
|
|
"num_tokens": 7854393.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.45957670211792,
|
|
"epoch": 3.401426388181355,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004261694901371807,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.1911183163523674,
|
|
"num_tokens": 7864891.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.563046598434449,
|
|
"epoch": 3.4065206316861945,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004258721150973155,
|
|
"loss": 5.2705,
|
|
"mean_token_accuracy": 0.18645353466272355,
|
|
"num_tokens": 7875170.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.472665500640869,
|
|
"epoch": 3.411614875191034,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004255742604379845,
|
|
"loss": 5.2798,
|
|
"mean_token_accuracy": 0.18504052013158798,
|
|
"num_tokens": 7887605.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.537009239196777,
|
|
"epoch": 3.4167091186958736,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004252759271060618,
|
|
"loss": 5.2824,
|
|
"mean_token_accuracy": 0.18396197408437728,
|
|
"num_tokens": 7899463.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.535813617706299,
|
|
"epoch": 3.421803362200713,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042497711604994314,
|
|
"loss": 5.3104,
|
|
"mean_token_accuracy": 0.1813850596547127,
|
|
"num_tokens": 7911897.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.484902381896973,
|
|
"epoch": 3.4268976057055527,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042467782821954306,
|
|
"loss": 5.2105,
|
|
"mean_token_accuracy": 0.19021005183458328,
|
|
"num_tokens": 7922687.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.539428281784057,
|
|
"epoch": 3.4319918492103922,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042437806456629154,
|
|
"loss": 5.2926,
|
|
"mean_token_accuracy": 0.1827096164226532,
|
|
"num_tokens": 7935604.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.477678012847901,
|
|
"epoch": 3.437086092715232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004240778260431311,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.18440396785736085,
|
|
"num_tokens": 7949083.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.5335972785949705,
|
|
"epoch": 3.4421803362200714,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004237771136045144,
|
|
"loss": 5.2517,
|
|
"mean_token_accuracy": 0.19293190836906432,
|
|
"num_tokens": 7962311.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.461599540710449,
|
|
"epoch": 3.447274579724911,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004234759282064001,
|
|
"loss": 5.2166,
|
|
"mean_token_accuracy": 0.19079147279262543,
|
|
"num_tokens": 7973531.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.527386426925659,
|
|
"epoch": 3.4523688232297505,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004231742708062508,
|
|
"loss": 5.1991,
|
|
"mean_token_accuracy": 0.1899975761771202,
|
|
"num_tokens": 7985927.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.538209342956543,
|
|
"epoch": 3.45746306673459,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004228721423630292,
|
|
"loss": 5.2882,
|
|
"mean_token_accuracy": 0.18557713478803634,
|
|
"num_tokens": 7997528.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.519944429397583,
|
|
"epoch": 3.4625573102394296,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004225695438371958,
|
|
"loss": 5.2284,
|
|
"mean_token_accuracy": 0.18910291343927382,
|
|
"num_tokens": 8010217.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.486494541168213,
|
|
"epoch": 3.467651553744269,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004222664761907052,
|
|
"loss": 5.2682,
|
|
"mean_token_accuracy": 0.1940535768866539,
|
|
"num_tokens": 8021261.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.503629064559936,
|
|
"epoch": 3.4727457972491087,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004219629403870035,
|
|
"loss": 5.3211,
|
|
"mean_token_accuracy": 0.1884480208158493,
|
|
"num_tokens": 8034044.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.487150049209594,
|
|
"epoch": 3.4778400407539483,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042165893739102504,
|
|
"loss": 5.2309,
|
|
"mean_token_accuracy": 0.1922961562871933,
|
|
"num_tokens": 8045312.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.542720365524292,
|
|
"epoch": 3.4829342842587874,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042135446816918927,
|
|
"loss": 5.2495,
|
|
"mean_token_accuracy": 0.18442296087741852,
|
|
"num_tokens": 8057800.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.569182395935059,
|
|
"epoch": 3.488028527763627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042104953368939785,
|
|
"loss": 5.326,
|
|
"mean_token_accuracy": 0.18522451668977738,
|
|
"num_tokens": 8069051.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.47875657081604,
|
|
"epoch": 3.4931227712684665,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004207441349210315,
|
|
"loss": 5.2557,
|
|
"mean_token_accuracy": 0.18691739588975906,
|
|
"num_tokens": 8080810.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.458851718902588,
|
|
"epoch": 3.498217014773306,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042043827283494667,
|
|
"loss": 5.2204,
|
|
"mean_token_accuracy": 0.1944128006696701,
|
|
"num_tokens": 8092421.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.545291328430176,
|
|
"epoch": 3.5033112582781456,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004201319484034731,
|
|
"loss": 5.2779,
|
|
"mean_token_accuracy": 0.18436517268419267,
|
|
"num_tokens": 8105384.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.558138799667359,
|
|
"epoch": 3.508405501782985,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041982516260040994,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.17795038670301438,
|
|
"num_tokens": 8118280.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.537533187866211,
|
|
"epoch": 3.5134997452878247,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041951791640102326,
|
|
"loss": 5.3024,
|
|
"mean_token_accuracy": 0.18330032527446746,
|
|
"num_tokens": 8129841.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.5610884666442875,
|
|
"epoch": 3.5185939887926643,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004192102107820426,
|
|
"loss": 5.2868,
|
|
"mean_token_accuracy": 0.1921681970357895,
|
|
"num_tokens": 8142222.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.548163032531738,
|
|
"epoch": 3.523688232297504,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041890204672165803,
|
|
"loss": 5.1956,
|
|
"mean_token_accuracy": 0.1938430428504944,
|
|
"num_tokens": 8152770.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.378242540359497,
|
|
"epoch": 3.5287824758023434,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00041859342519951696,
|
|
"loss": 5.2163,
|
|
"mean_token_accuracy": 0.18944479674100875,
|
|
"num_tokens": 8165599.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.560760068893432,
|
|
"epoch": 3.533876719307183,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004182843471967212,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.18509505242109298,
|
|
"num_tokens": 8177829.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.635271406173706,
|
|
"epoch": 3.5389709628120225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041797481369582355,
|
|
"loss": 5.3034,
|
|
"mean_token_accuracy": 0.18820751905441285,
|
|
"num_tokens": 8190148.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.435022068023682,
|
|
"epoch": 3.544065206316862,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041766482568082486,
|
|
"loss": 5.279,
|
|
"mean_token_accuracy": 0.1849372997879982,
|
|
"num_tokens": 8202072.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.570571660995483,
|
|
"epoch": 3.5491594498217016,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004173543841371709,
|
|
"loss": 5.2688,
|
|
"mean_token_accuracy": 0.18425412625074386,
|
|
"num_tokens": 8213460.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.507781028747559,
|
|
"epoch": 3.554253693326541,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004170434900517492,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.19250355064868926,
|
|
"num_tokens": 8225418.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.422589254379273,
|
|
"epoch": 3.5593479368313803,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041673214441288595,
|
|
"loss": 5.2053,
|
|
"mean_token_accuracy": 0.18687406778335572,
|
|
"num_tokens": 8236832.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.531172847747802,
|
|
"epoch": 3.5644421803362203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004164203482103426,
|
|
"loss": 5.2989,
|
|
"mean_token_accuracy": 0.18846114128828048,
|
|
"num_tokens": 8248655.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.526588010787964,
|
|
"epoch": 3.5695364238410594,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004161081024353132,
|
|
"loss": 5.2536,
|
|
"mean_token_accuracy": 0.18767817318439484,
|
|
"num_tokens": 8260171.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.493145513534546,
|
|
"epoch": 3.5746306673458994,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041579540808042096,
|
|
"loss": 5.2661,
|
|
"mean_token_accuracy": 0.1834352731704712,
|
|
"num_tokens": 8273402.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.517963409423828,
|
|
"epoch": 3.5797249108507385,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000415482266139715,
|
|
"loss": 5.3047,
|
|
"mean_token_accuracy": 0.18799168318510057,
|
|
"num_tokens": 8287594.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.529940986633301,
|
|
"epoch": 3.584819154355578,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004151686776086673,
|
|
"loss": 5.2722,
|
|
"mean_token_accuracy": 0.18521452099084854,
|
|
"num_tokens": 8299171.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.490778112411499,
|
|
"epoch": 3.5899133978604176,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041485464348416963,
|
|
"loss": 5.2135,
|
|
"mean_token_accuracy": 0.18747405111789703,
|
|
"num_tokens": 8310956.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.5366922378540036,
|
|
"epoch": 3.595007641365257,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041454016476453017,
|
|
"loss": 5.2866,
|
|
"mean_token_accuracy": 0.18736546039581298,
|
|
"num_tokens": 8322773.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.517085886001587,
|
|
"epoch": 3.6001018848700967,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041422524244947066,
|
|
"loss": 5.2656,
|
|
"mean_token_accuracy": 0.1847609758377075,
|
|
"num_tokens": 8334215.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.400420045852661,
|
|
"epoch": 3.6051961283749363,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004139098775401228,
|
|
"loss": 5.1586,
|
|
"mean_token_accuracy": 0.1995896577835083,
|
|
"num_tokens": 8345051.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.554336166381836,
|
|
"epoch": 3.610290371879776,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041359407103902556,
|
|
"loss": 5.3172,
|
|
"mean_token_accuracy": 0.18415218889713286,
|
|
"num_tokens": 8357730.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.510653972625732,
|
|
"epoch": 3.6153846153846154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004132778239501214,
|
|
"loss": 5.2766,
|
|
"mean_token_accuracy": 0.1828484430909157,
|
|
"num_tokens": 8369656.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.526478815078735,
|
|
"epoch": 3.620478858889455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041296113727875346,
|
|
"loss": 5.2775,
|
|
"mean_token_accuracy": 0.18837043941020964,
|
|
"num_tokens": 8381112.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.528132534027099,
|
|
"epoch": 3.6255731023942945,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041264401203166266,
|
|
"loss": 5.2472,
|
|
"mean_token_accuracy": 0.18709144443273545,
|
|
"num_tokens": 8393248.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.435088014602661,
|
|
"epoch": 3.630667345899134,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041232644921698357,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.19445168673992158,
|
|
"num_tokens": 8403976.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.613885545730591,
|
|
"epoch": 3.6357615894039736,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004120084498442425,
|
|
"loss": 5.327,
|
|
"mean_token_accuracy": 0.1806965798139572,
|
|
"num_tokens": 8416068.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.518724584579468,
|
|
"epoch": 3.640855832908813,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00041169001492435275,
|
|
"loss": 5.2546,
|
|
"mean_token_accuracy": 0.18656234294176102,
|
|
"num_tokens": 8428822.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.453029727935791,
|
|
"epoch": 3.6459500764136523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004113711454696128,
|
|
"loss": 5.2925,
|
|
"mean_token_accuracy": 0.18776277154684068,
|
|
"num_tokens": 8440888.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.556114768981933,
|
|
"epoch": 3.6510443199184923,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041105184249370236,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.1905710831284523,
|
|
"num_tokens": 8452734.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.41893572807312,
|
|
"epoch": 3.6561385634233314,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004107321070116792,
|
|
"loss": 5.2559,
|
|
"mean_token_accuracy": 0.18924230933189393,
|
|
"num_tokens": 8463472.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.473667621612549,
|
|
"epoch": 3.6612328069281714,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041041194003997616,
|
|
"loss": 5.247,
|
|
"mean_token_accuracy": 0.18323721140623092,
|
|
"num_tokens": 8475205.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.498313617706299,
|
|
"epoch": 3.6663270504330105,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041009134259639766,
|
|
"loss": 5.1774,
|
|
"mean_token_accuracy": 0.20673494637012482,
|
|
"num_tokens": 8486376.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.453317403793335,
|
|
"epoch": 3.67142129393785,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004097703157001166,
|
|
"loss": 5.2955,
|
|
"mean_token_accuracy": 0.18532535582780837,
|
|
"num_tokens": 8498507.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.494325923919678,
|
|
"epoch": 3.6765155374426897,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004094488603716712,
|
|
"loss": 5.2892,
|
|
"mean_token_accuracy": 0.19083812534809114,
|
|
"num_tokens": 8509675.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.481424236297608,
|
|
"epoch": 3.681609780947529,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004091269776329614,
|
|
"loss": 5.2451,
|
|
"mean_token_accuracy": 0.19099425822496413,
|
|
"num_tokens": 8520606.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.530824089050293,
|
|
"epoch": 3.6867040244523688,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004088046685072462,
|
|
"loss": 5.2972,
|
|
"mean_token_accuracy": 0.18068633526563643,
|
|
"num_tokens": 8532115.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.465677261352539,
|
|
"epoch": 3.6917982679572083,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040848193401914,
|
|
"loss": 5.1922,
|
|
"mean_token_accuracy": 0.19463919699192048,
|
|
"num_tokens": 8543875.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.503637218475342,
|
|
"epoch": 3.696892511462048,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040815877519460923,
|
|
"loss": 5.2962,
|
|
"mean_token_accuracy": 0.19162729978561402,
|
|
"num_tokens": 8555293.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.533040237426758,
|
|
"epoch": 3.7019867549668874,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040783519306096934,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.18606011420488358,
|
|
"num_tokens": 8567353.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.3949016571044925,
|
|
"epoch": 3.707080998471727,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004075111886468817,
|
|
"loss": 5.2358,
|
|
"mean_token_accuracy": 0.19684479683637618,
|
|
"num_tokens": 8578973.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.419895315170288,
|
|
"epoch": 3.7121752419765666,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040718676298234975,
|
|
"loss": 5.23,
|
|
"mean_token_accuracy": 0.1936740294098854,
|
|
"num_tokens": 8590770.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.570424747467041,
|
|
"epoch": 3.717269485481406,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040686191709871637,
|
|
"loss": 5.2883,
|
|
"mean_token_accuracy": 0.18951959908008575,
|
|
"num_tokens": 8602483.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.6009129047393795,
|
|
"epoch": 3.7223637289862457,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004065366520286601,
|
|
"loss": 5.372,
|
|
"mean_token_accuracy": 0.17853158861398696,
|
|
"num_tokens": 8613859.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.353788423538208,
|
|
"epoch": 3.7274579724910852,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004062109688061921,
|
|
"loss": 5.2008,
|
|
"mean_token_accuracy": 0.19954431653022767,
|
|
"num_tokens": 8625988.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.521049165725708,
|
|
"epoch": 3.7325522159959243,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040588486846665316,
|
|
"loss": 5.257,
|
|
"mean_token_accuracy": 0.18938526809215545,
|
|
"num_tokens": 8637777.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.508348798751831,
|
|
"epoch": 3.7376464595007644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004055583520467096,
|
|
"loss": 5.2907,
|
|
"mean_token_accuracy": 0.1842132478952408,
|
|
"num_tokens": 8648888.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.5662829875946045,
|
|
"epoch": 3.7427407030056035,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040523142058435064,
|
|
"loss": 5.3043,
|
|
"mean_token_accuracy": 0.18539980947971343,
|
|
"num_tokens": 8661530.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.507928133010864,
|
|
"epoch": 3.7478349465104435,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040490407511888495,
|
|
"loss": 5.2154,
|
|
"mean_token_accuracy": 0.19326558262109755,
|
|
"num_tokens": 8672246.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.428379726409912,
|
|
"epoch": 3.7529291900152826,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004045763166909374,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.1981903940439224,
|
|
"num_tokens": 8683621.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.490217685699463,
|
|
"epoch": 3.758023433520122,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004042481463424453,
|
|
"loss": 5.2968,
|
|
"mean_token_accuracy": 0.18723134100437164,
|
|
"num_tokens": 8695411.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.559792327880859,
|
|
"epoch": 3.7631176770249617,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004039195651166558,
|
|
"loss": 5.2805,
|
|
"mean_token_accuracy": 0.18421080857515335,
|
|
"num_tokens": 8706754.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.428022623062134,
|
|
"epoch": 3.7682119205298013,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004035905740581222,
|
|
"loss": 5.2856,
|
|
"mean_token_accuracy": 0.18863032758235931,
|
|
"num_tokens": 8718608.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.563587665557861,
|
|
"epoch": 3.773306164034641,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040326117421270055,
|
|
"loss": 5.3465,
|
|
"mean_token_accuracy": 0.182009194791317,
|
|
"num_tokens": 8731542.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.5590544700622555,
|
|
"epoch": 3.7784004075394804,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004029313666275463,
|
|
"loss": 5.3577,
|
|
"mean_token_accuracy": 0.17705416828393936,
|
|
"num_tokens": 8744531.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.485619449615479,
|
|
"epoch": 3.78349465104432,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004026011523511112,
|
|
"loss": 5.2893,
|
|
"mean_token_accuracy": 0.1867738962173462,
|
|
"num_tokens": 8756807.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.470218563079834,
|
|
"epoch": 3.7885888945491595,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040227053243314015,
|
|
"loss": 5.1525,
|
|
"mean_token_accuracy": 0.19636197984218598,
|
|
"num_tokens": 8767687.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.384046506881714,
|
|
"epoch": 3.793683138053999,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004019395079246672,
|
|
"loss": 5.2116,
|
|
"mean_token_accuracy": 0.1882733315229416,
|
|
"num_tokens": 8778662.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.5706493854522705,
|
|
"epoch": 3.7987773815588386,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040160807987801265,
|
|
"loss": 5.2849,
|
|
"mean_token_accuracy": 0.1872727856040001,
|
|
"num_tokens": 8791801.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.433224773406982,
|
|
"epoch": 3.803871625063678,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040127624934677966,
|
|
"loss": 5.2142,
|
|
"mean_token_accuracy": 0.19960336536169052,
|
|
"num_tokens": 8804404.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.505157136917115,
|
|
"epoch": 3.8089658685685177,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000400944017385851,
|
|
"loss": 5.3073,
|
|
"mean_token_accuracy": 0.18397554010152817,
|
|
"num_tokens": 8816862.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.5413895606994625,
|
|
"epoch": 3.8140601120733573,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040061138505138546,
|
|
"loss": 5.2291,
|
|
"mean_token_accuracy": 0.19314610958099365,
|
|
"num_tokens": 8827579.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.514079523086548,
|
|
"epoch": 3.8191543555781964,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004002783534008147,
|
|
"loss": 5.2434,
|
|
"mean_token_accuracy": 0.18283749222755433,
|
|
"num_tokens": 8839191.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.4834949493408205,
|
|
"epoch": 3.8242485990830364,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039994492349283966,
|
|
"loss": 5.2815,
|
|
"mean_token_accuracy": 0.19259392023086547,
|
|
"num_tokens": 8851936.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.471471643447876,
|
|
"epoch": 3.8293428425878755,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003996110963874274,
|
|
"loss": 5.2877,
|
|
"mean_token_accuracy": 0.1844303712248802,
|
|
"num_tokens": 8863575.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.507080698013306,
|
|
"epoch": 3.8344370860927155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039927687314580783,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.19102290272712708,
|
|
"num_tokens": 8874015.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.393535900115967,
|
|
"epoch": 3.8395313295975546,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039894225483047,
|
|
"loss": 5.156,
|
|
"mean_token_accuracy": 0.19848406314849854,
|
|
"num_tokens": 8885373.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.444007396697998,
|
|
"epoch": 3.844625573102394,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003986072425051587,
|
|
"loss": 5.1966,
|
|
"mean_token_accuracy": 0.18895271569490432,
|
|
"num_tokens": 8898238.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.456797409057617,
|
|
"epoch": 3.8497198166072337,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039827183723487156,
|
|
"loss": 5.3068,
|
|
"mean_token_accuracy": 0.18537504225969315,
|
|
"num_tokens": 8910832.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.40575795173645,
|
|
"epoch": 3.8548140601120733,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039793604008585536,
|
|
"loss": 5.217,
|
|
"mean_token_accuracy": 0.19796961396932602,
|
|
"num_tokens": 8923064.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.471729850769043,
|
|
"epoch": 3.859908303616913,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003975998521256026,
|
|
"loss": 5.2679,
|
|
"mean_token_accuracy": 0.19091691374778746,
|
|
"num_tokens": 8935871.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.522080945968628,
|
|
"epoch": 3.8650025471217524,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000397263274422848,
|
|
"loss": 5.3321,
|
|
"mean_token_accuracy": 0.18096788227558136,
|
|
"num_tokens": 8948170.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.56830620765686,
|
|
"epoch": 3.870096790626592,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003969263080475657,
|
|
"loss": 5.2632,
|
|
"mean_token_accuracy": 0.18650609850883484,
|
|
"num_tokens": 8959804.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.500931882858277,
|
|
"epoch": 3.8751910341314315,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039658895407096483,
|
|
"loss": 5.2689,
|
|
"mean_token_accuracy": 0.1917865604162216,
|
|
"num_tokens": 8971732.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.457354974746704,
|
|
"epoch": 3.880285277636271,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003962512135654871,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.19580780267715453,
|
|
"num_tokens": 8982084.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.4188981533050535,
|
|
"epoch": 3.8853795211411106,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039591308760480293,
|
|
"loss": 5.2458,
|
|
"mean_token_accuracy": 0.19003569185733796,
|
|
"num_tokens": 8993165.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.533962249755859,
|
|
"epoch": 3.89047376464595,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039557457726380803,
|
|
"loss": 5.2389,
|
|
"mean_token_accuracy": 0.19540747702121736,
|
|
"num_tokens": 9004496.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.519143867492676,
|
|
"epoch": 3.8955680081507897,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039523568361862007,
|
|
"loss": 5.3307,
|
|
"mean_token_accuracy": 0.1834275707602501,
|
|
"num_tokens": 9015643.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.449062395095825,
|
|
"epoch": 3.9006622516556293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039489640774657527,
|
|
"loss": 5.187,
|
|
"mean_token_accuracy": 0.19920356720685958,
|
|
"num_tokens": 9027469.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.6170331001281735,
|
|
"epoch": 3.9057564951604684,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039455675072622485,
|
|
"loss": 5.5079,
|
|
"mean_token_accuracy": 0.17070190608501434,
|
|
"num_tokens": 9041908.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.507436084747314,
|
|
"epoch": 3.9108507386653084,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003942167136373317,
|
|
"loss": 5.2776,
|
|
"mean_token_accuracy": 0.18503931015729905,
|
|
"num_tokens": 9053467.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.445768642425537,
|
|
"epoch": 3.9159449821701475,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003938762975608671,
|
|
"loss": 5.2531,
|
|
"mean_token_accuracy": 0.1980734497308731,
|
|
"num_tokens": 9064339.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.496502447128296,
|
|
"epoch": 3.9210392256749875,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003935355035790071,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.18633974492549896,
|
|
"num_tokens": 9076842.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.540559148788452,
|
|
"epoch": 3.9261334691798266,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039319433277512866,
|
|
"loss": 5.2849,
|
|
"mean_token_accuracy": 0.18864943832159042,
|
|
"num_tokens": 9089558.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.540275764465332,
|
|
"epoch": 3.931227712684666,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039285278623380746,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.1853303998708725,
|
|
"num_tokens": 9100266.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.5027618408203125,
|
|
"epoch": 3.9363219561895058,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039251086504081285,
|
|
"loss": 5.3275,
|
|
"mean_token_accuracy": 0.18472507745027542,
|
|
"num_tokens": 9112821.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.539356136322022,
|
|
"epoch": 3.9414161996943453,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039216857028310576,
|
|
"loss": 5.3268,
|
|
"mean_token_accuracy": 0.17800971120595932,
|
|
"num_tokens": 9123559.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.560427045822143,
|
|
"epoch": 3.946510443199185,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003918259030488342,
|
|
"loss": 5.2575,
|
|
"mean_token_accuracy": 0.18911080360412597,
|
|
"num_tokens": 9134707.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.497877311706543,
|
|
"epoch": 3.9516046867040244,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039148286442733073,
|
|
"loss": 5.1934,
|
|
"mean_token_accuracy": 0.19610525369644166,
|
|
"num_tokens": 9146101.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.3819365978240965,
|
|
"epoch": 3.956698930208864,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003911394555091083,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.1904640018939972,
|
|
"num_tokens": 9157394.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.495995759963989,
|
|
"epoch": 3.9617931737137035,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039079567738585703,
|
|
"loss": 5.1966,
|
|
"mean_token_accuracy": 0.1946175903081894,
|
|
"num_tokens": 9168851.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.477271842956543,
|
|
"epoch": 3.966887417218543,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003904515311504407,
|
|
"loss": 5.2716,
|
|
"mean_token_accuracy": 0.18859646171331407,
|
|
"num_tokens": 9180754.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.508467483520508,
|
|
"epoch": 3.9719816607233827,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039010701789689344,
|
|
"loss": 5.2127,
|
|
"mean_token_accuracy": 0.18830909878015517,
|
|
"num_tokens": 9191472.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.486693954467773,
|
|
"epoch": 3.977075904228222,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003897621387204161,
|
|
"loss": 5.2632,
|
|
"mean_token_accuracy": 0.18629163056612014,
|
|
"num_tokens": 9202761.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.511297369003296,
|
|
"epoch": 3.9821701477330618,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003894168947173727,
|
|
"loss": 5.2827,
|
|
"mean_token_accuracy": 0.18662455379962922,
|
|
"num_tokens": 9214980.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.583823394775391,
|
|
"epoch": 3.9872643912379013,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00038907128698528705,
|
|
"loss": 5.3886,
|
|
"mean_token_accuracy": 0.1783910095691681,
|
|
"num_tokens": 9227715.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.601547718048096,
|
|
"epoch": 3.9923586347427404,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003887253166228394,
|
|
"loss": 5.3232,
|
|
"mean_token_accuracy": 0.18276998847723008,
|
|
"num_tokens": 9239994.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.4391241550445555,
|
|
"epoch": 3.9974528782475804,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038837898472986247,
|
|
"loss": 5.2351,
|
|
"mean_token_accuracy": 0.1915659263730049,
|
|
"num_tokens": 9250950.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.434223333994548,
|
|
"epoch": 4.002037697401936,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038803229240733886,
|
|
"loss": 5.0621,
|
|
"mean_token_accuracy": 0.20407458808686998,
|
|
"num_tokens": 9260873.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.412366580963135,
|
|
"epoch": 4.007131940906775,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003876852407573962,
|
|
"loss": 4.8926,
|
|
"mean_token_accuracy": 0.2120436504483223,
|
|
"num_tokens": 9273272.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.461654424667358,
|
|
"epoch": 4.012226184411615,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000387337830883305,
|
|
"loss": 4.9881,
|
|
"mean_token_accuracy": 0.2071352407336235,
|
|
"num_tokens": 9284616.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.536989450454712,
|
|
"epoch": 4.017320427916454,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003869900638894743,
|
|
"loss": 4.9729,
|
|
"mean_token_accuracy": 0.20825144797563552,
|
|
"num_tokens": 9296284.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.453976440429687,
|
|
"epoch": 4.022414671421294,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00038664194088144867,
|
|
"loss": 4.9695,
|
|
"mean_token_accuracy": 0.20266242623329161,
|
|
"num_tokens": 9308654.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.381327533721924,
|
|
"epoch": 4.027508914926133,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000386293462965904,
|
|
"loss": 4.8779,
|
|
"mean_token_accuracy": 0.2097995936870575,
|
|
"num_tokens": 9320311.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.488311147689819,
|
|
"epoch": 4.032603158430973,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003859446312506449,
|
|
"loss": 4.9499,
|
|
"mean_token_accuracy": 0.2021094337105751,
|
|
"num_tokens": 9332859.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.345381546020508,
|
|
"epoch": 4.037697401935812,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003855954468446003,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.20052500218153,
|
|
"num_tokens": 9344365.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.407957124710083,
|
|
"epoch": 4.042791645440652,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038524591085782063,
|
|
"loss": 4.9077,
|
|
"mean_token_accuracy": 0.20996422171592713,
|
|
"num_tokens": 9357099.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.435814952850341,
|
|
"epoch": 4.047885888945491,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003848960244014738,
|
|
"loss": 4.9318,
|
|
"mean_token_accuracy": 0.20901981443166734,
|
|
"num_tokens": 9368764.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.3318421840667725,
|
|
"epoch": 4.052980132450331,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003845457885878421,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.21302445828914643,
|
|
"num_tokens": 9380853.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.4904468059539795,
|
|
"epoch": 4.0580743759551705,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003841952045303179,
|
|
"loss": 4.9644,
|
|
"mean_token_accuracy": 0.20485841929912568,
|
|
"num_tokens": 9393137.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.436367225646973,
|
|
"epoch": 4.0631686194600105,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003838442733434013,
|
|
"loss": 4.9444,
|
|
"mean_token_accuracy": 0.2037326067686081,
|
|
"num_tokens": 9404966.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.377862119674683,
|
|
"epoch": 4.06826286296485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038349299614269544,
|
|
"loss": 5.0225,
|
|
"mean_token_accuracy": 0.20129673629999162,
|
|
"num_tokens": 9417258.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.476395559310913,
|
|
"epoch": 4.07335710646969,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003831413740449037,
|
|
"loss": 4.9457,
|
|
"mean_token_accuracy": 0.21139460057020187,
|
|
"num_tokens": 9428281.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.07335710646969,
|
|
"eval_entropy": 5.255389784589226,
|
|
"eval_loss": 5.733760356903076,
|
|
"eval_mean_token_accuracy": 0.1741860475069211,
|
|
"eval_num_tokens": 9428281.0,
|
|
"eval_runtime": 2.7693,
|
|
"eval_samples_per_second": 1169.982,
|
|
"eval_steps_per_second": 146.248,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.337201166152954,
|
|
"epoch": 4.078451349974529,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003827894081678255,
|
|
"loss": 4.9252,
|
|
"mean_token_accuracy": 0.20863753110170363,
|
|
"num_tokens": 9440355.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.41506404876709,
|
|
"epoch": 4.083545593479369,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003824370996303538,
|
|
"loss": 4.934,
|
|
"mean_token_accuracy": 0.2108664706349373,
|
|
"num_tokens": 9451752.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.415864896774292,
|
|
"epoch": 4.088639836984208,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038208444955247005,
|
|
"loss": 4.9563,
|
|
"mean_token_accuracy": 0.21076004058122635,
|
|
"num_tokens": 9463794.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.317117691040039,
|
|
"epoch": 4.093734080489047,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003817314590552421,
|
|
"loss": 4.8797,
|
|
"mean_token_accuracy": 0.21737392097711564,
|
|
"num_tokens": 9476035.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.408023071289063,
|
|
"epoch": 4.098828323993887,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038137812926081986,
|
|
"loss": 4.9393,
|
|
"mean_token_accuracy": 0.20597162544727327,
|
|
"num_tokens": 9488590.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.438060331344604,
|
|
"epoch": 4.103922567498726,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003810244612924314,
|
|
"loss": 5.004,
|
|
"mean_token_accuracy": 0.20545936077833177,
|
|
"num_tokens": 9500211.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.499308633804321,
|
|
"epoch": 4.109016811003566,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003806704562743805,
|
|
"loss": 5.0571,
|
|
"mean_token_accuracy": 0.19622356742620467,
|
|
"num_tokens": 9513635.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.422359848022461,
|
|
"epoch": 4.114111054508405,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003803161153320418,
|
|
"loss": 4.9231,
|
|
"mean_token_accuracy": 0.21268320381641387,
|
|
"num_tokens": 9527466.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.388062858581543,
|
|
"epoch": 4.119205298013245,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037996143959185847,
|
|
"loss": 4.9151,
|
|
"mean_token_accuracy": 0.21365272998809814,
|
|
"num_tokens": 9537478.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.336142826080322,
|
|
"epoch": 4.124299541518084,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037960643018133743,
|
|
"loss": 4.9828,
|
|
"mean_token_accuracy": 0.20188651084899903,
|
|
"num_tokens": 9549645.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.403102445602417,
|
|
"epoch": 4.129393785022924,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037925108822904667,
|
|
"loss": 4.9129,
|
|
"mean_token_accuracy": 0.21041987836360931,
|
|
"num_tokens": 9561681.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.334452295303345,
|
|
"epoch": 4.134488028527763,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003788954148646112,
|
|
"loss": 4.891,
|
|
"mean_token_accuracy": 0.21443597823381425,
|
|
"num_tokens": 9573087.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.379920721054077,
|
|
"epoch": 4.139582272032603,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037853941121870964,
|
|
"loss": 4.9619,
|
|
"mean_token_accuracy": 0.20178107917308807,
|
|
"num_tokens": 9584662.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.370247793197632,
|
|
"epoch": 4.1446765155374425,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003781830784230704,
|
|
"loss": 4.9713,
|
|
"mean_token_accuracy": 0.21315461099147798,
|
|
"num_tokens": 9595609.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.3341302394866945,
|
|
"epoch": 4.1497707590422825,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00037782641761046856,
|
|
"loss": 4.8359,
|
|
"mean_token_accuracy": 0.2161322593688965,
|
|
"num_tokens": 9607424.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.289687824249268,
|
|
"epoch": 4.154865002547122,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037746942991472157,
|
|
"loss": 4.9275,
|
|
"mean_token_accuracy": 0.20669651478528978,
|
|
"num_tokens": 9618054.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.360912275314331,
|
|
"epoch": 4.159959246051962,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037711211647068637,
|
|
"loss": 5.0019,
|
|
"mean_token_accuracy": 0.2051767647266388,
|
|
"num_tokens": 9629936.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.41795711517334,
|
|
"epoch": 4.165053489556801,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037675447841425535,
|
|
"loss": 4.9866,
|
|
"mean_token_accuracy": 0.20483200997114182,
|
|
"num_tokens": 9642670.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.398531866073609,
|
|
"epoch": 4.170147733061641,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037639651688235276,
|
|
"loss": 4.9364,
|
|
"mean_token_accuracy": 0.2076996758580208,
|
|
"num_tokens": 9653800.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.403253126144409,
|
|
"epoch": 4.17524197656648,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003760382330129312,
|
|
"loss": 4.976,
|
|
"mean_token_accuracy": 0.2011588379740715,
|
|
"num_tokens": 9663814.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.37736701965332,
|
|
"epoch": 4.180336220071319,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000375679627944968,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.2080291137099266,
|
|
"num_tokens": 9676385.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.39991135597229,
|
|
"epoch": 4.185430463576159,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003753207028184617,
|
|
"loss": 4.9665,
|
|
"mean_token_accuracy": 0.20628130435943604,
|
|
"num_tokens": 9688225.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.342302036285401,
|
|
"epoch": 4.190524707080998,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000374961458774428,
|
|
"loss": 4.9338,
|
|
"mean_token_accuracy": 0.2135348066687584,
|
|
"num_tokens": 9702026.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.344960308074951,
|
|
"epoch": 4.195618950585838,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003746018969548966,
|
|
"loss": 4.9406,
|
|
"mean_token_accuracy": 0.2075305849313736,
|
|
"num_tokens": 9713991.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.409175825119019,
|
|
"epoch": 4.200713194090677,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003742420185029075,
|
|
"loss": 4.9505,
|
|
"mean_token_accuracy": 0.20534802079200745,
|
|
"num_tokens": 9724809.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.361985445022583,
|
|
"epoch": 4.205807437595517,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003738818245625072,
|
|
"loss": 5.0271,
|
|
"mean_token_accuracy": 0.200667367875576,
|
|
"num_tokens": 9737350.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.428904485702515,
|
|
"epoch": 4.210901681100356,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000373521316278745,
|
|
"loss": 5.0368,
|
|
"mean_token_accuracy": 0.19652751833200455,
|
|
"num_tokens": 9749092.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.422136926651001,
|
|
"epoch": 4.215995924605196,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003731604947976695,
|
|
"loss": 4.9988,
|
|
"mean_token_accuracy": 0.20478328317403793,
|
|
"num_tokens": 9763016.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.4074986457824705,
|
|
"epoch": 4.2210901681100355,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003727993612663253,
|
|
"loss": 4.982,
|
|
"mean_token_accuracy": 0.20139130353927612,
|
|
"num_tokens": 9775019.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.377883863449097,
|
|
"epoch": 4.2261844116148755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037243791683274865,
|
|
"loss": 4.9194,
|
|
"mean_token_accuracy": 0.21685723811388016,
|
|
"num_tokens": 9786385.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.369738388061523,
|
|
"epoch": 4.231278655119715,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003720761626459641,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.19842499494552612,
|
|
"num_tokens": 9797498.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.328862142562866,
|
|
"epoch": 4.236372898624555,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003717140998559812,
|
|
"loss": 4.9626,
|
|
"mean_token_accuracy": 0.21374436914920808,
|
|
"num_tokens": 9810654.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.349468135833741,
|
|
"epoch": 4.241467142129394,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00037135172961379035,
|
|
"loss": 4.8747,
|
|
"mean_token_accuracy": 0.21194127649068834,
|
|
"num_tokens": 9821169.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.381473207473755,
|
|
"epoch": 4.246561385634234,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003709890530713594,
|
|
"loss": 5.0091,
|
|
"mean_token_accuracy": 0.20810685604810714,
|
|
"num_tokens": 9832933.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.416640377044677,
|
|
"epoch": 4.251655629139073,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003706260713816298,
|
|
"loss": 5.0696,
|
|
"mean_token_accuracy": 0.205371955037117,
|
|
"num_tokens": 9844530.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.324978303909302,
|
|
"epoch": 4.256749872643912,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037026278569851325,
|
|
"loss": 4.9733,
|
|
"mean_token_accuracy": 0.20732166469097138,
|
|
"num_tokens": 9856056.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.473564386367798,
|
|
"epoch": 4.261844116148752,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003698991971768877,
|
|
"loss": 5.062,
|
|
"mean_token_accuracy": 0.1977630615234375,
|
|
"num_tokens": 9867534.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.408720684051514,
|
|
"epoch": 4.266938359653591,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003695353069725939,
|
|
"loss": 5.0167,
|
|
"mean_token_accuracy": 0.19899221658706664,
|
|
"num_tokens": 9879990.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.405464315414429,
|
|
"epoch": 4.272032603158431,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036917111624243143,
|
|
"loss": 5.0286,
|
|
"mean_token_accuracy": 0.19749937802553177,
|
|
"num_tokens": 9891835.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.383208322525024,
|
|
"epoch": 4.27712684666327,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036880662614415574,
|
|
"loss": 4.9145,
|
|
"mean_token_accuracy": 0.2133861303329468,
|
|
"num_tokens": 9903940.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.367089462280274,
|
|
"epoch": 4.28222109016811,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036844183783647333,
|
|
"loss": 5.0,
|
|
"mean_token_accuracy": 0.2019899994134903,
|
|
"num_tokens": 9915249.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.378032159805298,
|
|
"epoch": 4.287315333672949,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036807675247903915,
|
|
"loss": 4.9754,
|
|
"mean_token_accuracy": 0.20608662068843842,
|
|
"num_tokens": 9926840.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.402879858016968,
|
|
"epoch": 4.292409577177789,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003677113712324522,
|
|
"loss": 4.955,
|
|
"mean_token_accuracy": 0.1998097628355026,
|
|
"num_tokens": 9937156.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.359095764160156,
|
|
"epoch": 4.297503820682628,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036734569525825243,
|
|
"loss": 5.0023,
|
|
"mean_token_accuracy": 0.20681086033582688,
|
|
"num_tokens": 9949802.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.440337753295898,
|
|
"epoch": 4.302598064187468,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003669797257189163,
|
|
"loss": 5.0005,
|
|
"mean_token_accuracy": 0.20528244078159333,
|
|
"num_tokens": 9961988.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.340046167373657,
|
|
"epoch": 4.3076923076923075,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036661346377785397,
|
|
"loss": 4.9446,
|
|
"mean_token_accuracy": 0.20591541528701782,
|
|
"num_tokens": 9972834.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.359239101409912,
|
|
"epoch": 4.3127865511971475,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003662469105994046,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.21567050665616988,
|
|
"num_tokens": 9983579.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.424230575561523,
|
|
"epoch": 4.317880794701987,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036588006734883377,
|
|
"loss": 5.0669,
|
|
"mean_token_accuracy": 0.19964374005794525,
|
|
"num_tokens": 9995647.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.428868293762207,
|
|
"epoch": 4.322975038206827,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036551293519232865,
|
|
"loss": 5.0822,
|
|
"mean_token_accuracy": 0.2008001133799553,
|
|
"num_tokens": 10007551.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.3553540229797365,
|
|
"epoch": 4.328069281711666,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003651455152969954,
|
|
"loss": 4.9334,
|
|
"mean_token_accuracy": 0.20950321853160858,
|
|
"num_tokens": 10018032.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.419817590713501,
|
|
"epoch": 4.333163525216506,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036477780883085423,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.20247459560632705,
|
|
"num_tokens": 10029834.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.373254203796387,
|
|
"epoch": 4.338257768721345,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036440981696283695,
|
|
"loss": 5.0049,
|
|
"mean_token_accuracy": 0.2025153174996376,
|
|
"num_tokens": 10042035.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.393853187561035,
|
|
"epoch": 4.343352012226184,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036404154086278236,
|
|
"loss": 5.0361,
|
|
"mean_token_accuracy": 0.20168742835521697,
|
|
"num_tokens": 10053483.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.345834970474243,
|
|
"epoch": 4.348446255731024,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000363672981701433,
|
|
"loss": 4.9185,
|
|
"mean_token_accuracy": 0.21143516451120375,
|
|
"num_tokens": 10065082.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.444625902175903,
|
|
"epoch": 4.353540499235863,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00036330414065043087,
|
|
"loss": 4.939,
|
|
"mean_token_accuracy": 0.200561460852623,
|
|
"num_tokens": 10076555.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.386490774154663,
|
|
"epoch": 4.358634742740703,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036293501888231465,
|
|
"loss": 4.9555,
|
|
"mean_token_accuracy": 0.2065674066543579,
|
|
"num_tokens": 10087351.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.312270879745483,
|
|
"epoch": 4.363728986245542,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036256561757051493,
|
|
"loss": 4.9241,
|
|
"mean_token_accuracy": 0.20785604119300843,
|
|
"num_tokens": 10097685.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.3534657001495365,
|
|
"epoch": 4.368823229750382,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003621959378893514,
|
|
"loss": 5.01,
|
|
"mean_token_accuracy": 0.20611898750066757,
|
|
"num_tokens": 10108792.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.331159687042236,
|
|
"epoch": 4.373917473255221,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003618259810140281,
|
|
"loss": 5.0239,
|
|
"mean_token_accuracy": 0.19889876693487168,
|
|
"num_tokens": 10120303.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.44341926574707,
|
|
"epoch": 4.379011716760061,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036145574812063093,
|
|
"loss": 5.0814,
|
|
"mean_token_accuracy": 0.1973603919148445,
|
|
"num_tokens": 10132444.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.4096802234649655,
|
|
"epoch": 4.3841059602649,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003610852403861227,
|
|
"loss": 4.9399,
|
|
"mean_token_accuracy": 0.21335018426179886,
|
|
"num_tokens": 10144054.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.359166860580444,
|
|
"epoch": 4.38920020376974,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036071445898834043,
|
|
"loss": 4.9667,
|
|
"mean_token_accuracy": 0.20559958964586258,
|
|
"num_tokens": 10154741.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.352263021469116,
|
|
"epoch": 4.3942944472745795,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00036034340510599053,
|
|
"loss": 4.9562,
|
|
"mean_token_accuracy": 0.20368387699127197,
|
|
"num_tokens": 10166851.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.356820344924927,
|
|
"epoch": 4.3993886907794195,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035997207991864604,
|
|
"loss": 5.0064,
|
|
"mean_token_accuracy": 0.21012973338365554,
|
|
"num_tokens": 10178051.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.431884813308716,
|
|
"epoch": 4.404482934284259,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035960048460674244,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.20068919956684111,
|
|
"num_tokens": 10190175.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.469923734664917,
|
|
"epoch": 4.409577177789099,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035922862035157394,
|
|
"loss": 5.0911,
|
|
"mean_token_accuracy": 0.19503378719091416,
|
|
"num_tokens": 10201903.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.463651990890503,
|
|
"epoch": 4.414671421293938,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003588564883352895,
|
|
"loss": 5.1068,
|
|
"mean_token_accuracy": 0.19577255249023437,
|
|
"num_tokens": 10214690.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.404096889495849,
|
|
"epoch": 4.419765664798778,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003584840897408895,
|
|
"loss": 5.0399,
|
|
"mean_token_accuracy": 0.2034399539232254,
|
|
"num_tokens": 10227470.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.338438653945923,
|
|
"epoch": 4.424859908303617,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003581114257522216,
|
|
"loss": 4.9804,
|
|
"mean_token_accuracy": 0.2078890770673752,
|
|
"num_tokens": 10239027.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.43108549118042,
|
|
"epoch": 4.429954151808456,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035773849755397737,
|
|
"loss": 5.0425,
|
|
"mean_token_accuracy": 0.19425125569105148,
|
|
"num_tokens": 10250683.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.376652336120605,
|
|
"epoch": 4.435048395313296,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00035736530633168805,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.20214509218931198,
|
|
"num_tokens": 10261538.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.443653726577759,
|
|
"epoch": 4.440142638818136,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003569918532717212,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.19373902380466462,
|
|
"num_tokens": 10274232.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.394541311264038,
|
|
"epoch": 4.445236882322975,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035661813956127677,
|
|
"loss": 4.9735,
|
|
"mean_token_accuracy": 0.20597428679466248,
|
|
"num_tokens": 10285370.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.3227331161499025,
|
|
"epoch": 4.450331125827814,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035624416638838313,
|
|
"loss": 4.917,
|
|
"mean_token_accuracy": 0.21616034358739852,
|
|
"num_tokens": 10296825.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.37681827545166,
|
|
"epoch": 4.455425369332654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003558699349418937,
|
|
"loss": 4.9358,
|
|
"mean_token_accuracy": 0.21215563863515854,
|
|
"num_tokens": 10308881.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.335023212432861,
|
|
"epoch": 4.460519612837493,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003554954464114827,
|
|
"loss": 4.9592,
|
|
"mean_token_accuracy": 0.21119588166475295,
|
|
"num_tokens": 10321084.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.287936878204346,
|
|
"epoch": 4.465613856342333,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035512070198764185,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.2064596474170685,
|
|
"num_tokens": 10332148.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.444420337677002,
|
|
"epoch": 4.470708099847172,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035474570286167626,
|
|
"loss": 5.0058,
|
|
"mean_token_accuracy": 0.20302657634019852,
|
|
"num_tokens": 10343732.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.36590576171875,
|
|
"epoch": 4.475802343352012,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003543704502257006,
|
|
"loss": 4.9842,
|
|
"mean_token_accuracy": 0.2116266518831253,
|
|
"num_tokens": 10355661.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.39252405166626,
|
|
"epoch": 4.4808965868568515,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035399494527263566,
|
|
"loss": 4.9573,
|
|
"mean_token_accuracy": 0.2108900636434555,
|
|
"num_tokens": 10366218.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.456254720687866,
|
|
"epoch": 4.4859908303616916,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003536191891962043,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.19003964364528655,
|
|
"num_tokens": 10378691.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.412118244171142,
|
|
"epoch": 4.491085073866531,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003532431831909276,
|
|
"loss": 5.0986,
|
|
"mean_token_accuracy": 0.19658941328525542,
|
|
"num_tokens": 10390097.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.449026679992675,
|
|
"epoch": 4.496179317371371,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035286692845212114,
|
|
"loss": 4.9978,
|
|
"mean_token_accuracy": 0.20522923171520233,
|
|
"num_tokens": 10401516.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.282246446609497,
|
|
"epoch": 4.50127356087621,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035249042617589135,
|
|
"loss": 4.9612,
|
|
"mean_token_accuracy": 0.2135474979877472,
|
|
"num_tokens": 10413629.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.344106912612915,
|
|
"epoch": 4.50636780438105,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003521136775591315,
|
|
"loss": 5.0136,
|
|
"mean_token_accuracy": 0.19811417907476425,
|
|
"num_tokens": 10427451.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.355435228347778,
|
|
"epoch": 4.511462047885889,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000351736683799518,
|
|
"loss": 4.9749,
|
|
"mean_token_accuracy": 0.20652235746383668,
|
|
"num_tokens": 10438979.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.332303619384765,
|
|
"epoch": 4.516556291390728,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035135944609550654,
|
|
"loss": 4.9362,
|
|
"mean_token_accuracy": 0.21351114362478257,
|
|
"num_tokens": 10450208.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.439396762847901,
|
|
"epoch": 4.521650534895568,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035098196564632825,
|
|
"loss": 5.0063,
|
|
"mean_token_accuracy": 0.20526679307222367,
|
|
"num_tokens": 10461249.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.362916898727417,
|
|
"epoch": 4.526744778400408,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035060424365198603,
|
|
"loss": 4.947,
|
|
"mean_token_accuracy": 0.208339624106884,
|
|
"num_tokens": 10473222.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.355725479125977,
|
|
"epoch": 4.531839021905247,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003502262813132507,
|
|
"loss": 5.0651,
|
|
"mean_token_accuracy": 0.19462378025054933,
|
|
"num_tokens": 10485302.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.342121744155884,
|
|
"epoch": 4.536933265410086,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034984807983165695,
|
|
"loss": 5.0804,
|
|
"mean_token_accuracy": 0.20104012340307237,
|
|
"num_tokens": 10498377.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.467302322387695,
|
|
"epoch": 4.542027508914926,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003494696404094998,
|
|
"loss": 5.0332,
|
|
"mean_token_accuracy": 0.19976325780153276,
|
|
"num_tokens": 10510216.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.419010353088379,
|
|
"epoch": 4.547121752419765,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034909096424983073,
|
|
"loss": 4.9484,
|
|
"mean_token_accuracy": 0.2073616862297058,
|
|
"num_tokens": 10522152.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.321471881866455,
|
|
"epoch": 4.552215995924605,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003487120525564537,
|
|
"loss": 5.0237,
|
|
"mean_token_accuracy": 0.2027902528643608,
|
|
"num_tokens": 10534380.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.412211132049561,
|
|
"epoch": 4.5573102394294445,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003483329065339216,
|
|
"loss": 5.0915,
|
|
"mean_token_accuracy": 0.18930915296077727,
|
|
"num_tokens": 10545623.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.387938547134399,
|
|
"epoch": 4.5624044829342845,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003479535273875319,
|
|
"loss": 5.0516,
|
|
"mean_token_accuracy": 0.204361754655838,
|
|
"num_tokens": 10557626.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.346456480026245,
|
|
"epoch": 4.567498726439124,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034757391632332364,
|
|
"loss": 4.9711,
|
|
"mean_token_accuracy": 0.20607584118843078,
|
|
"num_tokens": 10569002.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.366862678527832,
|
|
"epoch": 4.572592969943964,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034719407454807274,
|
|
"loss": 4.9875,
|
|
"mean_token_accuracy": 0.20341406464576722,
|
|
"num_tokens": 10581458.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.316956472396851,
|
|
"epoch": 4.577687213448803,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003468140032692887,
|
|
"loss": 4.9688,
|
|
"mean_token_accuracy": 0.2061856985092163,
|
|
"num_tokens": 10593157.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.366600561141968,
|
|
"epoch": 4.582781456953643,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003464337036952105,
|
|
"loss": 4.975,
|
|
"mean_token_accuracy": 0.2066730245947838,
|
|
"num_tokens": 10603504.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.475948190689087,
|
|
"epoch": 4.587875700458482,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034605317703480313,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.19639609307050704,
|
|
"num_tokens": 10617186.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.418831253051758,
|
|
"epoch": 4.592969943963322,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003456724244977533,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.20278005301952362,
|
|
"num_tokens": 10629871.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.370974922180176,
|
|
"epoch": 4.598064187468161,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003452914472944656,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.20510291904211045,
|
|
"num_tokens": 10641461.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.3840250968933105,
|
|
"epoch": 4.603158430973,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003449102466360591,
|
|
"loss": 4.9714,
|
|
"mean_token_accuracy": 0.20671036541461946,
|
|
"num_tokens": 10652811.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.341228246688843,
|
|
"epoch": 4.60825267447784,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 4.9529,
|
|
"mean_token_accuracy": 0.20559901148080825,
|
|
"num_tokens": 10663624.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.3920900344848635,
|
|
"epoch": 4.61334691798268,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034414717980191375,
|
|
"loss": 5.0122,
|
|
"mean_token_accuracy": 0.202510803937912,
|
|
"num_tokens": 10675317.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.39328966140747,
|
|
"epoch": 4.618441161487519,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034376531605194915,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.21023057401180267,
|
|
"num_tokens": 10686343.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.482343912124634,
|
|
"epoch": 4.623535404992358,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003433832336984068,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.1920938163995743,
|
|
"num_tokens": 10697688.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.399043703079224,
|
|
"epoch": 4.628629648497198,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003430009339559189,
|
|
"loss": 4.9364,
|
|
"mean_token_accuracy": 0.2078985556960106,
|
|
"num_tokens": 10707840.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.3677583694458,
|
|
"epoch": 4.633723892002037,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003426184180398087,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.20405758917331696,
|
|
"num_tokens": 10719702.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.302867889404297,
|
|
"epoch": 4.638818135506877,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003422356871660868,
|
|
"loss": 4.9034,
|
|
"mean_token_accuracy": 0.2075626865029335,
|
|
"num_tokens": 10731040.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.397806835174561,
|
|
"epoch": 4.6439123790117165,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034185274255144705,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.2052369236946106,
|
|
"num_tokens": 10743009.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.427628803253174,
|
|
"epoch": 4.6490066225165565,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003414695854132629,
|
|
"loss": 5.0788,
|
|
"mean_token_accuracy": 0.20144866704940795,
|
|
"num_tokens": 10754030.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.35647873878479,
|
|
"epoch": 4.654100866021396,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003410862169695831,
|
|
"loss": 4.9793,
|
|
"mean_token_accuracy": 0.2110179990530014,
|
|
"num_tokens": 10765512.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.366422033309936,
|
|
"epoch": 4.659195109526236,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003407026384391284,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.1981634333729744,
|
|
"num_tokens": 10778694.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.365662097930908,
|
|
"epoch": 4.664289353031075,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003403188510412874,
|
|
"loss": 4.9134,
|
|
"mean_token_accuracy": 0.21068077236413957,
|
|
"num_tokens": 10789458.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.396241283416748,
|
|
"epoch": 4.669383596535915,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033993485599611266,
|
|
"loss": 4.9839,
|
|
"mean_token_accuracy": 0.20414211899042128,
|
|
"num_tokens": 10801932.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.266009902954101,
|
|
"epoch": 4.674477840040754,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033955065452431684,
|
|
"loss": 4.999,
|
|
"mean_token_accuracy": 0.2077878937125206,
|
|
"num_tokens": 10814907.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.351363611221314,
|
|
"epoch": 4.679572083545594,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00033916624784726865,
|
|
"loss": 4.967,
|
|
"mean_token_accuracy": 0.20493731200695037,
|
|
"num_tokens": 10825707.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.340922832489014,
|
|
"epoch": 4.684666327050433,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033878163718698945,
|
|
"loss": 5.0408,
|
|
"mean_token_accuracy": 0.20442576408386232,
|
|
"num_tokens": 10839498.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.463293170928955,
|
|
"epoch": 4.689760570555272,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033839682376614875,
|
|
"loss": 5.073,
|
|
"mean_token_accuracy": 0.19963130801916124,
|
|
"num_tokens": 10851507.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.419545745849609,
|
|
"epoch": 4.694854814060112,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033801180880806115,
|
|
"loss": 5.0357,
|
|
"mean_token_accuracy": 0.20111733675003052,
|
|
"num_tokens": 10863437.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.398361158370972,
|
|
"epoch": 4.699949057564952,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003376265935366812,
|
|
"loss": 4.9963,
|
|
"mean_token_accuracy": 0.20897321850061418,
|
|
"num_tokens": 10875499.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.323073959350586,
|
|
"epoch": 4.705043301069791,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003372411791766008,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.20254020243883133,
|
|
"num_tokens": 10886006.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.344968843460083,
|
|
"epoch": 4.71013754457463,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033685556695304446,
|
|
"loss": 4.998,
|
|
"mean_token_accuracy": 0.19904941469430923,
|
|
"num_tokens": 10899258.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.373987913131714,
|
|
"epoch": 4.71523178807947,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033646975809186597,
|
|
"loss": 5.0364,
|
|
"mean_token_accuracy": 0.2035851627588272,
|
|
"num_tokens": 10910672.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.4716414451599125,
|
|
"epoch": 4.720326031584309,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003360837538195439,
|
|
"loss": 5.0825,
|
|
"mean_token_accuracy": 0.20119217038154602,
|
|
"num_tokens": 10923087.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.394886493682861,
|
|
"epoch": 4.725420275089149,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003356975553631782,
|
|
"loss": 5.0201,
|
|
"mean_token_accuracy": 0.20163317322731017,
|
|
"num_tokens": 10935715.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.416485977172852,
|
|
"epoch": 4.7305145185939885,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033531116395048614,
|
|
"loss": 5.0193,
|
|
"mean_token_accuracy": 0.20252590328454972,
|
|
"num_tokens": 10946016.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.378958749771118,
|
|
"epoch": 4.7356087620988285,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003349245808097985,
|
|
"loss": 4.9851,
|
|
"mean_token_accuracy": 0.2063085362315178,
|
|
"num_tokens": 10958733.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.4327140808105465,
|
|
"epoch": 4.740703005603668,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003345378071700553,
|
|
"loss": 5.0338,
|
|
"mean_token_accuracy": 0.20945581942796707,
|
|
"num_tokens": 10970519.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.336383867263794,
|
|
"epoch": 4.745797249108508,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003341508442608022,
|
|
"loss": 4.9766,
|
|
"mean_token_accuracy": 0.20345241725444793,
|
|
"num_tokens": 10981420.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.3883708953857425,
|
|
"epoch": 4.750891492613347,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003337636933121867,
|
|
"loss": 5.1284,
|
|
"mean_token_accuracy": 0.19337846636772155,
|
|
"num_tokens": 10993735.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.372046089172363,
|
|
"epoch": 4.755985736118187,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003333763555549541,
|
|
"loss": 4.9205,
|
|
"mean_token_accuracy": 0.21437881886959076,
|
|
"num_tokens": 11004077.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.362741422653198,
|
|
"epoch": 4.761079979623026,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003329888322204434,
|
|
"loss": 5.0112,
|
|
"mean_token_accuracy": 0.20396677255630494,
|
|
"num_tokens": 11015478.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.420085382461548,
|
|
"epoch": 4.766174223127866,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033260112454058344,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.2018229126930237,
|
|
"num_tokens": 11026785.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.370578193664551,
|
|
"epoch": 4.771268466632705,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033221323374788946,
|
|
"loss": 5.0517,
|
|
"mean_token_accuracy": 0.19932416528463365,
|
|
"num_tokens": 11037616.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.3910938739776615,
|
|
"epoch": 4.776362710137544,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033182516107545864,
|
|
"loss": 5.1028,
|
|
"mean_token_accuracy": 0.1966820850968361,
|
|
"num_tokens": 11049565.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.374528980255127,
|
|
"epoch": 4.781456953642384,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033143690775696606,
|
|
"loss": 4.9556,
|
|
"mean_token_accuracy": 0.20718374103307724,
|
|
"num_tokens": 11061456.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.233138942718506,
|
|
"epoch": 4.786551197147224,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003310484750266616,
|
|
"loss": 4.9397,
|
|
"mean_token_accuracy": 0.21421614438295364,
|
|
"num_tokens": 11074206.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.362055110931396,
|
|
"epoch": 4.791645440652063,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033065986411936506,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.20574764907360077,
|
|
"num_tokens": 11086550.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.4614965438842775,
|
|
"epoch": 4.796739684156902,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003302710762704629,
|
|
"loss": 5.0682,
|
|
"mean_token_accuracy": 0.1974300980567932,
|
|
"num_tokens": 11098616.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.3571137428283695,
|
|
"epoch": 4.801833927661742,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000329882112715904,
|
|
"loss": 5.0244,
|
|
"mean_token_accuracy": 0.2021840915083885,
|
|
"num_tokens": 11109015.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.3717857837677006,
|
|
"epoch": 4.806928171166581,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032949297469219563,
|
|
"loss": 4.9701,
|
|
"mean_token_accuracy": 0.20520471334457396,
|
|
"num_tokens": 11120694.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.350765562057495,
|
|
"epoch": 4.8120224146714214,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032910366343639997,
|
|
"loss": 5.0187,
|
|
"mean_token_accuracy": 0.19979261457920075,
|
|
"num_tokens": 11131646.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.434678840637207,
|
|
"epoch": 4.817116658176261,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032871418018612993,
|
|
"loss": 5.0967,
|
|
"mean_token_accuracy": 0.19378001540899276,
|
|
"num_tokens": 11143786.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.412089681625366,
|
|
"epoch": 4.822210901681101,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003283245261795448,
|
|
"loss": 5.0069,
|
|
"mean_token_accuracy": 0.201558156311512,
|
|
"num_tokens": 11155505.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.327121305465698,
|
|
"epoch": 4.82730514518594,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003279347026553472,
|
|
"loss": 5.0075,
|
|
"mean_token_accuracy": 0.20318959802389144,
|
|
"num_tokens": 11167719.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.445400857925415,
|
|
"epoch": 4.83239938869078,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003275447108527782,
|
|
"loss": 5.0886,
|
|
"mean_token_accuracy": 0.19832610636949538,
|
|
"num_tokens": 11180289.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.394275236129761,
|
|
"epoch": 4.837493632195619,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003271545520116141,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.20345490723848342,
|
|
"num_tokens": 11192657.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.386234378814697,
|
|
"epoch": 4.842587875700459,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003267642273721621,
|
|
"loss": 5.0332,
|
|
"mean_token_accuracy": 0.2061567097902298,
|
|
"num_tokens": 11203822.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.429087495803833,
|
|
"epoch": 4.847682119205298,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032637373817525657,
|
|
"loss": 5.0715,
|
|
"mean_token_accuracy": 0.20295971482992173,
|
|
"num_tokens": 11215504.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.373113822937012,
|
|
"epoch": 4.852776362710138,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003259830856622547,
|
|
"loss": 4.9748,
|
|
"mean_token_accuracy": 0.21102215349674225,
|
|
"num_tokens": 11225905.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.3430687427520756,
|
|
"epoch": 4.857870606214977,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032559227107503344,
|
|
"loss": 4.9735,
|
|
"mean_token_accuracy": 0.20944474190473555,
|
|
"num_tokens": 11237396.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.426400279998779,
|
|
"epoch": 4.862964849719816,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032520129565598435,
|
|
"loss": 5.0437,
|
|
"mean_token_accuracy": 0.20344182550907136,
|
|
"num_tokens": 11247693.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.319186544418335,
|
|
"epoch": 4.868059093224656,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003248101606480106,
|
|
"loss": 4.9535,
|
|
"mean_token_accuracy": 0.19995065331459044,
|
|
"num_tokens": 11257716.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.331370544433594,
|
|
"epoch": 4.873153336729496,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003244188672945228,
|
|
"loss": 4.9662,
|
|
"mean_token_accuracy": 0.20698574781417847,
|
|
"num_tokens": 11269427.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.384437704086304,
|
|
"epoch": 4.878247580234335,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032402741683943456,
|
|
"loss": 4.9914,
|
|
"mean_token_accuracy": 0.2130946159362793,
|
|
"num_tokens": 11281758.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.3566474437713625,
|
|
"epoch": 4.883341823739174,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003236358105271593,
|
|
"loss": 5.0317,
|
|
"mean_token_accuracy": 0.20372913181781768,
|
|
"num_tokens": 11292926.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.420602655410766,
|
|
"epoch": 4.888436067244014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003232440496026055,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.20072022527456285,
|
|
"num_tokens": 11306022.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.396597385406494,
|
|
"epoch": 4.8935303107488535,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003228521353111736,
|
|
"loss": 5.002,
|
|
"mean_token_accuracy": 0.20532903373241423,
|
|
"num_tokens": 11318252.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.3756482124328615,
|
|
"epoch": 4.8986245542536935,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003224600688987513,
|
|
"loss": 4.9851,
|
|
"mean_token_accuracy": 0.20986895859241486,
|
|
"num_tokens": 11330604.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.371840476989746,
|
|
"epoch": 4.903718797758533,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003220678516117099,
|
|
"loss": 5.0312,
|
|
"mean_token_accuracy": 0.2040150299668312,
|
|
"num_tokens": 11343020.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.424312019348145,
|
|
"epoch": 4.908813041263373,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032167548469690045,
|
|
"loss": 5.0723,
|
|
"mean_token_accuracy": 0.19578463733196258,
|
|
"num_tokens": 11354727.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.4078779220581055,
|
|
"epoch": 4.913907284768212,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003212829694016496,
|
|
"loss": 5.059,
|
|
"mean_token_accuracy": 0.2014452636241913,
|
|
"num_tokens": 11366437.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.303188467025757,
|
|
"epoch": 4.919001528273052,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003208903069737557,
|
|
"loss": 4.957,
|
|
"mean_token_accuracy": 0.20959644615650178,
|
|
"num_tokens": 11379877.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.357510900497436,
|
|
"epoch": 4.924095771777891,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003204974986614848,
|
|
"loss": 4.9817,
|
|
"mean_token_accuracy": 0.2076667383313179,
|
|
"num_tokens": 11392113.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.474664258956909,
|
|
"epoch": 4.929190015282731,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032010454571356677,
|
|
"loss": 5.0634,
|
|
"mean_token_accuracy": 0.1995212987065315,
|
|
"num_tokens": 11404790.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.36910080909729,
|
|
"epoch": 4.93428425878757,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003197114493791913,
|
|
"loss": 5.0107,
|
|
"mean_token_accuracy": 0.20567060858011246,
|
|
"num_tokens": 11417515.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.2877452850341795,
|
|
"epoch": 4.93937850229241,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031931821090800365,
|
|
"loss": 4.9575,
|
|
"mean_token_accuracy": 0.21871894001960754,
|
|
"num_tokens": 11430188.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.288412046432495,
|
|
"epoch": 4.944472745797249,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031892483155010145,
|
|
"loss": 4.8911,
|
|
"mean_token_accuracy": 0.2162867844104767,
|
|
"num_tokens": 11442648.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.327475452423096,
|
|
"epoch": 4.949566989302088,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003185313125560295,
|
|
"loss": 4.9742,
|
|
"mean_token_accuracy": 0.21348634064197541,
|
|
"num_tokens": 11453587.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.378320693969727,
|
|
"epoch": 4.954661232806928,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031813765517677727,
|
|
"loss": 4.9112,
|
|
"mean_token_accuracy": 0.20764364749193193,
|
|
"num_tokens": 11464476.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.371028995513916,
|
|
"epoch": 4.959755476311768,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003177438606637736,
|
|
"loss": 5.0369,
|
|
"mean_token_accuracy": 0.19999713748693465,
|
|
"num_tokens": 11476907.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.261577987670899,
|
|
"epoch": 4.964849719816607,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003173499302688834,
|
|
"loss": 4.9307,
|
|
"mean_token_accuracy": 0.20986434668302537,
|
|
"num_tokens": 11488392.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.353446578979492,
|
|
"epoch": 4.969943963321446,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031695586524440364,
|
|
"loss": 4.9631,
|
|
"mean_token_accuracy": 0.20578839629888535,
|
|
"num_tokens": 11499882.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.379978370666504,
|
|
"epoch": 4.975038206826286,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003165616668430592,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.1971799150109291,
|
|
"num_tokens": 11511669.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.364163064956665,
|
|
"epoch": 4.9801324503311255,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031616733631799907,
|
|
"loss": 4.9762,
|
|
"mean_token_accuracy": 0.20628045499324799,
|
|
"num_tokens": 11523850.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.213493680953979,
|
|
"epoch": 4.9852266938359655,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031577287492279217,
|
|
"loss": 4.9387,
|
|
"mean_token_accuracy": 0.2121466189622879,
|
|
"num_tokens": 11535295.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.442234706878662,
|
|
"epoch": 4.990320937340805,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003153782839114233,
|
|
"loss": 5.0711,
|
|
"mean_token_accuracy": 0.19502150416374206,
|
|
"num_tokens": 11546304.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.4286956787109375,
|
|
"epoch": 4.995415180845645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031498356453828967,
|
|
"loss": 5.1205,
|
|
"mean_token_accuracy": 0.19723245799541472,
|
|
"num_tokens": 11558886.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.412975364261204,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003145887180581963,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.19745104180441964,
|
|
"num_tokens": 11570380.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.367984294891357,
|
|
"epoch": 5.005094243504839,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003141937457263523,
|
|
"loss": 4.6992,
|
|
"mean_token_accuracy": 0.22854919880628585,
|
|
"num_tokens": 11581170.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.404342842102051,
|
|
"epoch": 5.010188487009679,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003137986487983668,
|
|
"loss": 4.8395,
|
|
"mean_token_accuracy": 0.21525125354528427,
|
|
"num_tokens": 11594407.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.325063037872314,
|
|
"epoch": 5.015282730514518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031340342853024533,
|
|
"loss": 4.7559,
|
|
"mean_token_accuracy": 0.22644097954034806,
|
|
"num_tokens": 11607159.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.3462242603302,
|
|
"epoch": 5.020376974019358,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003130080861783851,
|
|
"loss": 4.7191,
|
|
"mean_token_accuracy": 0.2272819072008133,
|
|
"num_tokens": 11617692.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.356867456436158,
|
|
"epoch": 5.025471217524197,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003126126229995717,
|
|
"loss": 4.8048,
|
|
"mean_token_accuracy": 0.21876227855682373,
|
|
"num_tokens": 11629679.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.281940364837647,
|
|
"epoch": 5.030565461029037,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003122170402509747,
|
|
"loss": 4.685,
|
|
"mean_token_accuracy": 0.22766069918870926,
|
|
"num_tokens": 11642507.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.320653438568115,
|
|
"epoch": 5.0356597045338765,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003118213391901439,
|
|
"loss": 4.7236,
|
|
"mean_token_accuracy": 0.22492515295743942,
|
|
"num_tokens": 11654709.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.275740909576416,
|
|
"epoch": 5.0407539480387165,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000311425521075005,
|
|
"loss": 4.7453,
|
|
"mean_token_accuracy": 0.22185977399349213,
|
|
"num_tokens": 11666266.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.296951675415039,
|
|
"epoch": 5.045848191543556,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003110295871638561,
|
|
"loss": 4.6784,
|
|
"mean_token_accuracy": 0.22748540788888932,
|
|
"num_tokens": 11677844.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.326102685928345,
|
|
"epoch": 5.050942435048396,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003106335387153632,
|
|
"loss": 4.7917,
|
|
"mean_token_accuracy": 0.21802024245262147,
|
|
"num_tokens": 11689202.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.368196582794189,
|
|
"epoch": 5.056036678553235,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003102373769885565,
|
|
"loss": 4.767,
|
|
"mean_token_accuracy": 0.21702695935964583,
|
|
"num_tokens": 11700155.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.264258098602295,
|
|
"epoch": 5.061130922058075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003098411032428262,
|
|
"loss": 4.7055,
|
|
"mean_token_accuracy": 0.22915479242801667,
|
|
"num_tokens": 11712463.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.310811185836792,
|
|
"epoch": 5.066225165562914,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00030944471873791876,
|
|
"loss": 4.793,
|
|
"mean_token_accuracy": 0.21323116421699523,
|
|
"num_tokens": 11724373.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.371287059783936,
|
|
"epoch": 5.071319409067754,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003090482247339326,
|
|
"loss": 4.7648,
|
|
"mean_token_accuracy": 0.22466946244239808,
|
|
"num_tokens": 11736394.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.286581802368164,
|
|
"epoch": 5.076413652572593,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003086516224913144,
|
|
"loss": 4.7225,
|
|
"mean_token_accuracy": 0.22794594168663024,
|
|
"num_tokens": 11747895.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.356030035018921,
|
|
"epoch": 5.081507896077433,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003082549132708548,
|
|
"loss": 4.7729,
|
|
"mean_token_accuracy": 0.22193437367677687,
|
|
"num_tokens": 11761287.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.260648250579834,
|
|
"epoch": 5.086602139582272,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030785809833368456,
|
|
"loss": 4.7366,
|
|
"mean_token_accuracy": 0.22918733060359955,
|
|
"num_tokens": 11773033.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.286873769760132,
|
|
"epoch": 5.091696383087111,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030746117894127047,
|
|
"loss": 4.7647,
|
|
"mean_token_accuracy": 0.22577632814645768,
|
|
"num_tokens": 11785101.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.091696383087111,
|
|
"eval_entropy": 5.17792983820409,
|
|
"eval_loss": 5.675128936767578,
|
|
"eval_mean_token_accuracy": 0.17964577845953128,
|
|
"eval_num_tokens": 11785101.0,
|
|
"eval_runtime": 2.8569,
|
|
"eval_samples_per_second": 1134.115,
|
|
"eval_steps_per_second": 141.764,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"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": 1.87854426432e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|