4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.036678553234845,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691933536529541,
|
|
"epoch": 0.005094243504839531,
|
|
"grad_norm": 12.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8617,
|
|
"mean_token_accuracy": 7.283321465365589e-05,
|
|
"num_tokens": 12441.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69195785522461,
|
|
"epoch": 0.010188487009679063,
|
|
"grad_norm": 13.0625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7886,
|
|
"mean_token_accuracy": 8.764241938479245e-05,
|
|
"num_tokens": 23925.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691264724731445,
|
|
"epoch": 0.015282730514518594,
|
|
"grad_norm": 9.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5167,
|
|
"mean_token_accuracy": 0.023061814554966985,
|
|
"num_tokens": 33958.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.679726791381835,
|
|
"epoch": 0.020376974019358125,
|
|
"grad_norm": 5.84375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1946,
|
|
"mean_token_accuracy": 0.043799113109707834,
|
|
"num_tokens": 46188.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.613013648986817,
|
|
"epoch": 0.02547121752419766,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8889,
|
|
"mean_token_accuracy": 0.04678206667304039,
|
|
"num_tokens": 58080.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.596247959136964,
|
|
"epoch": 0.030565461029037188,
|
|
"grad_norm": 3.171875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7705,
|
|
"mean_token_accuracy": 0.03978040032088757,
|
|
"num_tokens": 72446.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.60612449645996,
|
|
"epoch": 0.03565970453387672,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6958,
|
|
"mean_token_accuracy": 0.043326153233647344,
|
|
"num_tokens": 85522.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.586102485656738,
|
|
"epoch": 0.04075394803871625,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6414,
|
|
"mean_token_accuracy": 0.039182791113853456,
|
|
"num_tokens": 96970.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.593886470794677,
|
|
"epoch": 0.04584819154355578,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6151,
|
|
"mean_token_accuracy": 0.04260034691542387,
|
|
"num_tokens": 108011.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.594364643096924,
|
|
"epoch": 0.05094243504839532,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.528,
|
|
"mean_token_accuracy": 0.04013910107314587,
|
|
"num_tokens": 119673.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.586376857757568,
|
|
"epoch": 0.056036678553234846,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5493,
|
|
"mean_token_accuracy": 0.0417574979364872,
|
|
"num_tokens": 133041.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.560707283020019,
|
|
"epoch": 0.061130922058074376,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4204,
|
|
"mean_token_accuracy": 0.042354696244001386,
|
|
"num_tokens": 143467.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.495057964324952,
|
|
"epoch": 0.06622516556291391,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3327,
|
|
"mean_token_accuracy": 0.05373127907514572,
|
|
"num_tokens": 154745.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.405490398406982,
|
|
"epoch": 0.07131940906775344,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2218,
|
|
"mean_token_accuracy": 0.053240276873111725,
|
|
"num_tokens": 165526.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.435882568359375,
|
|
"epoch": 0.07641365257259297,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1812,
|
|
"mean_token_accuracy": 0.054624059423804286,
|
|
"num_tokens": 177735.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.438795280456542,
|
|
"epoch": 0.0815078960774325,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0299,
|
|
"mean_token_accuracy": 0.06236044131219387,
|
|
"num_tokens": 189877.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.311788082122803,
|
|
"epoch": 0.08660213958227203,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0334,
|
|
"mean_token_accuracy": 0.058551866188645364,
|
|
"num_tokens": 201740.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.396924591064453,
|
|
"epoch": 0.09169638308711156,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9084,
|
|
"mean_token_accuracy": 0.05712463036179542,
|
|
"num_tokens": 213385.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.31311674118042,
|
|
"epoch": 0.09679062659195109,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8263,
|
|
"mean_token_accuracy": 0.06144661605358124,
|
|
"num_tokens": 225501.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.3136625289917,
|
|
"epoch": 0.10188487009679063,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7332,
|
|
"mean_token_accuracy": 0.06830099895596504,
|
|
"num_tokens": 239512.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.148342895507813,
|
|
"epoch": 0.10697911360163016,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6085,
|
|
"mean_token_accuracy": 0.06609614193439484,
|
|
"num_tokens": 252261.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.107837390899657,
|
|
"epoch": 0.11207335710646969,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4972,
|
|
"mean_token_accuracy": 0.06363288164138795,
|
|
"num_tokens": 263754.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.950089359283448,
|
|
"epoch": 0.11716760061130922,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.377,
|
|
"mean_token_accuracy": 0.07086736857891082,
|
|
"num_tokens": 274953.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.918815803527831,
|
|
"epoch": 0.12226184411614875,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2772,
|
|
"mean_token_accuracy": 0.07252867110073566,
|
|
"num_tokens": 284844.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.788673591613769,
|
|
"epoch": 0.1273560876209883,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1819,
|
|
"mean_token_accuracy": 0.07298735603690147,
|
|
"num_tokens": 296833.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.649177837371827,
|
|
"epoch": 0.13245033112582782,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0676,
|
|
"mean_token_accuracy": 0.07574639841914177,
|
|
"num_tokens": 308528.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.483296966552734,
|
|
"epoch": 0.13754457463066735,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9082,
|
|
"mean_token_accuracy": 0.07914942651987075,
|
|
"num_tokens": 320599.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.217451667785644,
|
|
"epoch": 0.14263881813550688,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8002,
|
|
"mean_token_accuracy": 0.07902043983340264,
|
|
"num_tokens": 332264.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.995352077484132,
|
|
"epoch": 0.1477330616403464,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7372,
|
|
"mean_token_accuracy": 0.08456770777702331,
|
|
"num_tokens": 344233.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.8959490776062,
|
|
"epoch": 0.15282730514518594,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.7246,
|
|
"mean_token_accuracy": 0.08076966851949692,
|
|
"num_tokens": 356200.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.64002866744995,
|
|
"epoch": 0.15792154865002547,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7279,
|
|
"mean_token_accuracy": 0.07943359091877937,
|
|
"num_tokens": 369791.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.485534954071046,
|
|
"epoch": 0.163015792154865,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.506,
|
|
"mean_token_accuracy": 0.08976328000426292,
|
|
"num_tokens": 381127.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.262951564788818,
|
|
"epoch": 0.16811003565970453,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6364,
|
|
"mean_token_accuracy": 0.0788124717772007,
|
|
"num_tokens": 393941.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.217638492584229,
|
|
"epoch": 0.17320427916454406,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4653,
|
|
"mean_token_accuracy": 0.08866559341549873,
|
|
"num_tokens": 404883.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.12268590927124,
|
|
"epoch": 0.1782985226693836,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5067,
|
|
"mean_token_accuracy": 0.0853340208530426,
|
|
"num_tokens": 417961.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.083796834945678,
|
|
"epoch": 0.18339276617422312,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4617,
|
|
"mean_token_accuracy": 0.08598975539207458,
|
|
"num_tokens": 430076.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.989891767501831,
|
|
"epoch": 0.18848700967906265,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4382,
|
|
"mean_token_accuracy": 0.08668889105319977,
|
|
"num_tokens": 442091.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.941371583938599,
|
|
"epoch": 0.19358125318390218,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4342,
|
|
"mean_token_accuracy": 0.08515545055270195,
|
|
"num_tokens": 454272.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.948634958267212,
|
|
"epoch": 0.1986754966887417,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.368,
|
|
"mean_token_accuracy": 0.07814025804400444,
|
|
"num_tokens": 465237.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.879694080352783,
|
|
"epoch": 0.20376974019358127,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.08662542253732682,
|
|
"num_tokens": 476842.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.8720118522644045,
|
|
"epoch": 0.2088639836984208,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4507,
|
|
"mean_token_accuracy": 0.0870475135743618,
|
|
"num_tokens": 489198.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.757378482818604,
|
|
"epoch": 0.21395822720326033,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3925,
|
|
"mean_token_accuracy": 0.089558956772089,
|
|
"num_tokens": 500320.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.914836311340332,
|
|
"epoch": 0.21905247070809986,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2584,
|
|
"mean_token_accuracy": 0.0958659790456295,
|
|
"num_tokens": 512117.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.7668720245361325,
|
|
"epoch": 0.22414671421293939,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.39,
|
|
"mean_token_accuracy": 0.08571948632597923,
|
|
"num_tokens": 524106.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.755286931991577,
|
|
"epoch": 0.22924095771777891,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3119,
|
|
"mean_token_accuracy": 0.08375829830765724,
|
|
"num_tokens": 535815.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.795973873138427,
|
|
"epoch": 0.23433520122261844,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3265,
|
|
"mean_token_accuracy": 0.08689820542931556,
|
|
"num_tokens": 548217.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.734084177017212,
|
|
"epoch": 0.23942944472745797,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3159,
|
|
"mean_token_accuracy": 0.09363964423537255,
|
|
"num_tokens": 559218.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.692209482192993,
|
|
"epoch": 0.2445236882322975,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2849,
|
|
"mean_token_accuracy": 0.08573080748319625,
|
|
"num_tokens": 571021.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.653146457672119,
|
|
"epoch": 0.24961793173713703,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3651,
|
|
"mean_token_accuracy": 0.08879496678709983,
|
|
"num_tokens": 582023.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.691405630111694,
|
|
"epoch": 0.2547121752419766,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.288,
|
|
"mean_token_accuracy": 0.09284965097904205,
|
|
"num_tokens": 594938.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.677452039718628,
|
|
"epoch": 0.2598064187468161,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2142,
|
|
"mean_token_accuracy": 0.09495285749435425,
|
|
"num_tokens": 606978.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.555742835998535,
|
|
"epoch": 0.26490066225165565,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.224,
|
|
"mean_token_accuracy": 0.09364139512181283,
|
|
"num_tokens": 617653.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.647906541824341,
|
|
"epoch": 0.26999490575649515,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.209,
|
|
"mean_token_accuracy": 0.08683466166257858,
|
|
"num_tokens": 628807.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.582596492767334,
|
|
"epoch": 0.2750891492613347,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2779,
|
|
"mean_token_accuracy": 0.088419159501791,
|
|
"num_tokens": 639512.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.654620885848999,
|
|
"epoch": 0.2801833927661742,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2792,
|
|
"mean_token_accuracy": 0.08684752136468887,
|
|
"num_tokens": 652491.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.6056561946868895,
|
|
"epoch": 0.28527763627101377,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1743,
|
|
"mean_token_accuracy": 0.09496726170182228,
|
|
"num_tokens": 664152.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.49957766532898,
|
|
"epoch": 0.29037187977585327,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.2412,
|
|
"mean_token_accuracy": 0.08711678311228752,
|
|
"num_tokens": 676774.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.589282417297364,
|
|
"epoch": 0.2954661232806928,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.256,
|
|
"mean_token_accuracy": 0.09429504796862602,
|
|
"num_tokens": 689459.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.601702690124512,
|
|
"epoch": 0.30056036678553233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1178,
|
|
"mean_token_accuracy": 0.1071195811033249,
|
|
"num_tokens": 700338.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.5238135814666744,
|
|
"epoch": 0.3056546102903719,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1756,
|
|
"mean_token_accuracy": 0.10010429248213767,
|
|
"num_tokens": 711764.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.4707728862762455,
|
|
"epoch": 0.3107488537952114,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1665,
|
|
"mean_token_accuracy": 0.09938320368528367,
|
|
"num_tokens": 725232.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.504101991653442,
|
|
"epoch": 0.31584309730005095,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.126,
|
|
"mean_token_accuracy": 0.09905346557497978,
|
|
"num_tokens": 737036.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.535156774520874,
|
|
"epoch": 0.32093734080489045,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.09537394866347312,
|
|
"num_tokens": 747844.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.532956981658936,
|
|
"epoch": 0.32603158430973,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.196,
|
|
"mean_token_accuracy": 0.09172512367367744,
|
|
"num_tokens": 760730.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.569301605224609,
|
|
"epoch": 0.33112582781456956,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1939,
|
|
"mean_token_accuracy": 0.08795616179704666,
|
|
"num_tokens": 773020.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.3828143119812015,
|
|
"epoch": 0.33622007131940906,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1709,
|
|
"mean_token_accuracy": 0.09435797706246377,
|
|
"num_tokens": 786337.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.46736159324646,
|
|
"epoch": 0.3413143148242486,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1946,
|
|
"mean_token_accuracy": 0.09482537731528282,
|
|
"num_tokens": 798490.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.484088563919068,
|
|
"epoch": 0.3464085583290881,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0577,
|
|
"mean_token_accuracy": 0.09521348923444747,
|
|
"num_tokens": 810448.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.479926538467407,
|
|
"epoch": 0.3515028018339277,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1475,
|
|
"mean_token_accuracy": 0.09660323038697242,
|
|
"num_tokens": 821443.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.534337854385376,
|
|
"epoch": 0.3565970453387672,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1518,
|
|
"mean_token_accuracy": 0.09190995395183563,
|
|
"num_tokens": 833935.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.371308040618897,
|
|
"epoch": 0.36169128884360674,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.1277,
|
|
"mean_token_accuracy": 0.09297349452972412,
|
|
"num_tokens": 845343.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.360507678985596,
|
|
"epoch": 0.36678553234844624,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0661,
|
|
"mean_token_accuracy": 0.10084544345736504,
|
|
"num_tokens": 856683.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.551660776138306,
|
|
"epoch": 0.3718797758532858,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1496,
|
|
"mean_token_accuracy": 0.09621937721967697,
|
|
"num_tokens": 869190.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.396897411346435,
|
|
"epoch": 0.3769740193581253,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1107,
|
|
"mean_token_accuracy": 0.09819533750414848,
|
|
"num_tokens": 880811.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.379519033432007,
|
|
"epoch": 0.38206826286296486,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.062,
|
|
"mean_token_accuracy": 0.09250850528478623,
|
|
"num_tokens": 892559.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.351970911026001,
|
|
"epoch": 0.38716250636780436,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9836,
|
|
"mean_token_accuracy": 0.09942182153463364,
|
|
"num_tokens": 903602.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.36213903427124,
|
|
"epoch": 0.3922567498726439,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0794,
|
|
"mean_token_accuracy": 0.09812569916248322,
|
|
"num_tokens": 915021.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.400866937637329,
|
|
"epoch": 0.3973509933774834,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1524,
|
|
"mean_token_accuracy": 0.09627576470375061,
|
|
"num_tokens": 927084.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.322605895996094,
|
|
"epoch": 0.402445236882323,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0569,
|
|
"mean_token_accuracy": 0.09742958173155784,
|
|
"num_tokens": 939900.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.3560632228851315,
|
|
"epoch": 0.40753948038716253,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1292,
|
|
"mean_token_accuracy": 0.08907266408205032,
|
|
"num_tokens": 953956.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.4070006847381595,
|
|
"epoch": 0.41263372389200204,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0604,
|
|
"mean_token_accuracy": 0.1046214260160923,
|
|
"num_tokens": 966526.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.261282110214234,
|
|
"epoch": 0.4177279673968416,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9502,
|
|
"mean_token_accuracy": 0.09846476316452027,
|
|
"num_tokens": 977889.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.307093572616577,
|
|
"epoch": 0.4228222109016811,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0391,
|
|
"mean_token_accuracy": 0.09500454217195511,
|
|
"num_tokens": 989737.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.332889413833618,
|
|
"epoch": 0.42791645440652065,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0727,
|
|
"mean_token_accuracy": 0.09839839786291123,
|
|
"num_tokens": 1000820.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.271477127075196,
|
|
"epoch": 0.43301069791136015,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9583,
|
|
"mean_token_accuracy": 0.10580145046114922,
|
|
"num_tokens": 1012156.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.290036249160766,
|
|
"epoch": 0.4381049414161997,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8937,
|
|
"mean_token_accuracy": 0.10478519052267074,
|
|
"num_tokens": 1022825.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.220993757247925,
|
|
"epoch": 0.4431991849210392,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9762,
|
|
"mean_token_accuracy": 0.10158220753073692,
|
|
"num_tokens": 1034725.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.262880039215088,
|
|
"epoch": 0.44829342842587877,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9638,
|
|
"mean_token_accuracy": 0.09944465085864067,
|
|
"num_tokens": 1045927.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.248775577545166,
|
|
"epoch": 0.4533876719307183,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9828,
|
|
"mean_token_accuracy": 0.09821130633354187,
|
|
"num_tokens": 1057943.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.3402995586395265,
|
|
"epoch": 0.45848191543555783,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0581,
|
|
"mean_token_accuracy": 0.09455323070287705,
|
|
"num_tokens": 1068949.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.2297731876373295,
|
|
"epoch": 0.46357615894039733,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.932,
|
|
"mean_token_accuracy": 0.1041158229112625,
|
|
"num_tokens": 1081298.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.228163337707519,
|
|
"epoch": 0.4686704024452369,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9816,
|
|
"mean_token_accuracy": 0.10278591811656952,
|
|
"num_tokens": 1092522.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.093091297149658,
|
|
"epoch": 0.4737646459500764,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8957,
|
|
"mean_token_accuracy": 0.1022353433072567,
|
|
"num_tokens": 1104133.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.173185110092163,
|
|
"epoch": 0.47885888945491595,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8605,
|
|
"mean_token_accuracy": 0.1085218146443367,
|
|
"num_tokens": 1116057.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.134055089950562,
|
|
"epoch": 0.48395313295975545,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.09869935512542724,
|
|
"num_tokens": 1128148.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.2346595287322994,
|
|
"epoch": 0.489047376464595,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9666,
|
|
"mean_token_accuracy": 0.09960033446550369,
|
|
"num_tokens": 1140104.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.173930263519287,
|
|
"epoch": 0.49414161996943456,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8951,
|
|
"mean_token_accuracy": 0.10067171081900597,
|
|
"num_tokens": 1152711.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.140958261489868,
|
|
"epoch": 0.49923586347427407,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8955,
|
|
"mean_token_accuracy": 0.10290815010666847,
|
|
"num_tokens": 1164223.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.230376243591309,
|
|
"epoch": 0.5043301069791136,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8962,
|
|
"mean_token_accuracy": 0.10572774037718773,
|
|
"num_tokens": 1174990.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.084908771514892,
|
|
"epoch": 0.5094243504839532,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8948,
|
|
"mean_token_accuracy": 0.10063612461090088,
|
|
"num_tokens": 1187561.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5094243504839532,
|
|
"eval_entropy": 7.262294474943184,
|
|
"eval_loss": 6.917758464813232,
|
|
"eval_mean_token_accuracy": 0.10530804161120344,
|
|
"eval_num_tokens": 1187561.0,
|
|
"eval_runtime": 2.7331,
|
|
"eval_samples_per_second": 1185.468,
|
|
"eval_steps_per_second": 148.183,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.168519735336304,
|
|
"epoch": 0.5145185939887926,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8901,
|
|
"mean_token_accuracy": 0.10450750216841698,
|
|
"num_tokens": 1199108.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.162225580215454,
|
|
"epoch": 0.5196128374936322,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9639,
|
|
"mean_token_accuracy": 0.10146603956818581,
|
|
"num_tokens": 1211944.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.139475297927857,
|
|
"epoch": 0.5247070809984717,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8997,
|
|
"mean_token_accuracy": 0.09549630656838418,
|
|
"num_tokens": 1223044.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.140495634078979,
|
|
"epoch": 0.5298013245033113,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.923,
|
|
"mean_token_accuracy": 0.10165961533784866,
|
|
"num_tokens": 1235161.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.138057518005371,
|
|
"epoch": 0.5348955680081507,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9507,
|
|
"mean_token_accuracy": 0.10160766541957855,
|
|
"num_tokens": 1247732.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.0351362228393555,
|
|
"epoch": 0.5399898115129903,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7937,
|
|
"mean_token_accuracy": 0.10965558364987374,
|
|
"num_tokens": 1259349.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.072154712677002,
|
|
"epoch": 0.5450840550178299,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.76,
|
|
"mean_token_accuracy": 0.11131609529256821,
|
|
"num_tokens": 1269992.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.97647762298584,
|
|
"epoch": 0.5501782985226694,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8285,
|
|
"mean_token_accuracy": 0.11009727045893669,
|
|
"num_tokens": 1281502.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.11279182434082,
|
|
"epoch": 0.5552725420275089,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7735,
|
|
"mean_token_accuracy": 0.10252350270748138,
|
|
"num_tokens": 1293880.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.021842575073242,
|
|
"epoch": 0.5603667855323484,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8787,
|
|
"mean_token_accuracy": 0.10112191140651702,
|
|
"num_tokens": 1306022.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.022417259216309,
|
|
"epoch": 0.565461029037188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8458,
|
|
"mean_token_accuracy": 0.10293983742594719,
|
|
"num_tokens": 1317939.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.911152935028076,
|
|
"epoch": 0.5705552725420275,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6947,
|
|
"mean_token_accuracy": 0.11798072904348374,
|
|
"num_tokens": 1329917.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.976852607727051,
|
|
"epoch": 0.5756495160468671,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8079,
|
|
"mean_token_accuracy": 0.11052986085414887,
|
|
"num_tokens": 1341733.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.096958923339844,
|
|
"epoch": 0.5807437595517065,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.746,
|
|
"mean_token_accuracy": 0.1133323922753334,
|
|
"num_tokens": 1352454.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.072584772109986,
|
|
"epoch": 0.5858380030565461,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.9179,
|
|
"mean_token_accuracy": 0.10664055198431015,
|
|
"num_tokens": 1363878.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.035218572616577,
|
|
"epoch": 0.5909322465613857,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8355,
|
|
"mean_token_accuracy": 0.09741255789995193,
|
|
"num_tokens": 1375159.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.033307790756226,
|
|
"epoch": 0.5960264900662252,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8531,
|
|
"mean_token_accuracy": 0.10392893180251121,
|
|
"num_tokens": 1386646.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.03128490447998,
|
|
"epoch": 0.6011207335710647,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7635,
|
|
"mean_token_accuracy": 0.11585776656866073,
|
|
"num_tokens": 1398020.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.957646226882934,
|
|
"epoch": 0.6062149770759042,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7512,
|
|
"mean_token_accuracy": 0.11048697158694268,
|
|
"num_tokens": 1408209.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.011190795898438,
|
|
"epoch": 0.6113092205807438,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8055,
|
|
"mean_token_accuracy": 0.1033931739628315,
|
|
"num_tokens": 1420391.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.984811115264892,
|
|
"epoch": 0.6164034640855833,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.7731,
|
|
"mean_token_accuracy": 0.10607712045311928,
|
|
"num_tokens": 1432250.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.982828664779663,
|
|
"epoch": 0.6214977075904228,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7448,
|
|
"mean_token_accuracy": 0.10987704768776893,
|
|
"num_tokens": 1444823.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.871470260620117,
|
|
"epoch": 0.6265919510952623,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8034,
|
|
"mean_token_accuracy": 0.10774814710021019,
|
|
"num_tokens": 1455997.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.018206071853638,
|
|
"epoch": 0.6316861946001019,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7605,
|
|
"mean_token_accuracy": 0.11087881028652191,
|
|
"num_tokens": 1467707.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.00219144821167,
|
|
"epoch": 0.6367804381049414,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.789,
|
|
"mean_token_accuracy": 0.10458588898181916,
|
|
"num_tokens": 1477920.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.005989694595337,
|
|
"epoch": 0.6418746816097809,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.8028,
|
|
"mean_token_accuracy": 0.10765366032719612,
|
|
"num_tokens": 1488176.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.924004459381104,
|
|
"epoch": 0.6469689251146205,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7612,
|
|
"mean_token_accuracy": 0.10841824859380722,
|
|
"num_tokens": 1499018.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.021401691436767,
|
|
"epoch": 0.65206316861946,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.10359535291790962,
|
|
"num_tokens": 1510759.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.915233039855957,
|
|
"epoch": 0.6571574121242996,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7829,
|
|
"mean_token_accuracy": 0.10922421887516975,
|
|
"num_tokens": 1523028.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.967010641098023,
|
|
"epoch": 0.6622516556291391,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7709,
|
|
"mean_token_accuracy": 0.11187901347875595,
|
|
"num_tokens": 1533505.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.995042371749878,
|
|
"epoch": 0.6673458991339786,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.8077,
|
|
"mean_token_accuracy": 0.11219450533390045,
|
|
"num_tokens": 1545366.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.975549745559692,
|
|
"epoch": 0.6724401426388181,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6689,
|
|
"mean_token_accuracy": 0.11294186711311341,
|
|
"num_tokens": 1557976.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.857566213607788,
|
|
"epoch": 0.6775343861436577,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7871,
|
|
"mean_token_accuracy": 0.10875996574759483,
|
|
"num_tokens": 1569822.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.903141641616822,
|
|
"epoch": 0.6826286296484972,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6939,
|
|
"mean_token_accuracy": 0.11107023805379868,
|
|
"num_tokens": 1580485.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.974818754196167,
|
|
"epoch": 0.6877228731533367,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.739,
|
|
"mean_token_accuracy": 0.106230578571558,
|
|
"num_tokens": 1591865.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.857933616638183,
|
|
"epoch": 0.6928171166581762,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6193,
|
|
"mean_token_accuracy": 0.11482649818062782,
|
|
"num_tokens": 1604580.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.869587802886963,
|
|
"epoch": 0.6979113601630158,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7781,
|
|
"mean_token_accuracy": 0.10850742682814599,
|
|
"num_tokens": 1615772.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.902220678329468,
|
|
"epoch": 0.7030056036678554,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.667,
|
|
"mean_token_accuracy": 0.11990016177296639,
|
|
"num_tokens": 1626965.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.808654689788819,
|
|
"epoch": 0.7080998471726948,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.7007,
|
|
"mean_token_accuracy": 0.11295107081532478,
|
|
"num_tokens": 1638865.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.851603651046753,
|
|
"epoch": 0.7131940906775344,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.673,
|
|
"mean_token_accuracy": 0.1173295795917511,
|
|
"num_tokens": 1649200.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.836373996734619,
|
|
"epoch": 0.7182883341823739,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6496,
|
|
"mean_token_accuracy": 0.11917404681444169,
|
|
"num_tokens": 1660726.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.016081094741821,
|
|
"epoch": 0.7233825776872135,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6981,
|
|
"mean_token_accuracy": 0.11023402512073517,
|
|
"num_tokens": 1672101.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.692198133468628,
|
|
"epoch": 0.7284768211920529,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.647,
|
|
"mean_token_accuracy": 0.11462761536240577,
|
|
"num_tokens": 1683560.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.85313572883606,
|
|
"epoch": 0.7335710646968925,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6876,
|
|
"mean_token_accuracy": 0.11544215381145477,
|
|
"num_tokens": 1694084.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.773692846298218,
|
|
"epoch": 0.738665308201732,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.11710209399461746,
|
|
"num_tokens": 1706686.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.84816575050354,
|
|
"epoch": 0.7437595517065716,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.7517,
|
|
"mean_token_accuracy": 0.11291900128126145,
|
|
"num_tokens": 1719661.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.915406131744385,
|
|
"epoch": 0.7488537952114112,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6261,
|
|
"mean_token_accuracy": 0.1130788080394268,
|
|
"num_tokens": 1730521.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.753418588638306,
|
|
"epoch": 0.7539480387162506,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7005,
|
|
"mean_token_accuracy": 0.11622855588793754,
|
|
"num_tokens": 1742251.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.651693344116211,
|
|
"epoch": 0.7590422822210902,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5389,
|
|
"mean_token_accuracy": 0.11808115169405937,
|
|
"num_tokens": 1752871.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.831832647323608,
|
|
"epoch": 0.7641365257259297,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7215,
|
|
"mean_token_accuracy": 0.11175637394189834,
|
|
"num_tokens": 1765690.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.77952823638916,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6015,
|
|
"mean_token_accuracy": 0.1158766008913517,
|
|
"num_tokens": 1778398.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.785519695281982,
|
|
"epoch": 0.7743250127356087,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6874,
|
|
"mean_token_accuracy": 0.10822736173868179,
|
|
"num_tokens": 1791496.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.903592872619629,
|
|
"epoch": 0.7794192562404483,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.683,
|
|
"mean_token_accuracy": 0.10840215981006622,
|
|
"num_tokens": 1803152.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.712223196029663,
|
|
"epoch": 0.7845134997452878,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6069,
|
|
"mean_token_accuracy": 0.11322797611355781,
|
|
"num_tokens": 1814775.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.67111029624939,
|
|
"epoch": 0.7896077432501274,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.11556519940495491,
|
|
"num_tokens": 1826068.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.748933267593384,
|
|
"epoch": 0.7947019867549668,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5929,
|
|
"mean_token_accuracy": 0.1146427683532238,
|
|
"num_tokens": 1838115.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.834074974060059,
|
|
"epoch": 0.7997962302598064,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.692,
|
|
"mean_token_accuracy": 0.11109071299433708,
|
|
"num_tokens": 1850670.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.708985233306885,
|
|
"epoch": 0.804890473764646,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6679,
|
|
"mean_token_accuracy": 0.11313713565468789,
|
|
"num_tokens": 1863500.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.828090238571167,
|
|
"epoch": 0.8099847172694855,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.11277235373854637,
|
|
"num_tokens": 1875840.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.7314472675323485,
|
|
"epoch": 0.8150789607743251,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6138,
|
|
"mean_token_accuracy": 0.11551589146256447,
|
|
"num_tokens": 1887265.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.735307455062866,
|
|
"epoch": 0.8201732042791645,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4903,
|
|
"mean_token_accuracy": 0.1243679828941822,
|
|
"num_tokens": 1898550.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.597352409362793,
|
|
"epoch": 0.8252674477840041,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4267,
|
|
"mean_token_accuracy": 0.12633467689156533,
|
|
"num_tokens": 1908990.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.649030351638794,
|
|
"epoch": 0.8303616912888436,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5239,
|
|
"mean_token_accuracy": 0.1194791056215763,
|
|
"num_tokens": 1920109.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.675817346572876,
|
|
"epoch": 0.8354559347936832,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5882,
|
|
"mean_token_accuracy": 0.1158361554145813,
|
|
"num_tokens": 1930936.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.7580108642578125,
|
|
"epoch": 0.8405501782985226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.11999981254339218,
|
|
"num_tokens": 1940968.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.54950590133667,
|
|
"epoch": 0.8456444218033622,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5245,
|
|
"mean_token_accuracy": 0.12037713453173637,
|
|
"num_tokens": 1952955.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.697274208068848,
|
|
"epoch": 0.8507386653082017,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.6222,
|
|
"mean_token_accuracy": 0.11852857768535614,
|
|
"num_tokens": 1966023.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.72812933921814,
|
|
"epoch": 0.8558329088130413,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5554,
|
|
"mean_token_accuracy": 0.12165540158748626,
|
|
"num_tokens": 1978036.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.567156553268433,
|
|
"epoch": 0.8609271523178808,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.529,
|
|
"mean_token_accuracy": 0.11831365376710892,
|
|
"num_tokens": 1988870.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.692898988723755,
|
|
"epoch": 0.8660213958227203,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.6079,
|
|
"mean_token_accuracy": 0.12303238585591317,
|
|
"num_tokens": 2000010.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.666892623901367,
|
|
"epoch": 0.8711156393275599,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5306,
|
|
"mean_token_accuracy": 0.11588020399212837,
|
|
"num_tokens": 2011127.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.691326379776001,
|
|
"epoch": 0.8762098828323994,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6304,
|
|
"mean_token_accuracy": 0.1136388897895813,
|
|
"num_tokens": 2023483.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.642430973052979,
|
|
"epoch": 0.8813041263372389,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6096,
|
|
"mean_token_accuracy": 0.1140765056014061,
|
|
"num_tokens": 2034699.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.6784350872039795,
|
|
"epoch": 0.8863983698420784,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.557,
|
|
"mean_token_accuracy": 0.11363079845905304,
|
|
"num_tokens": 2046849.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.733535099029541,
|
|
"epoch": 0.891492613346918,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6706,
|
|
"mean_token_accuracy": 0.11693223789334298,
|
|
"num_tokens": 2060535.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.701980686187744,
|
|
"epoch": 0.8965868568517575,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5288,
|
|
"mean_token_accuracy": 0.12081804201006889,
|
|
"num_tokens": 2072576.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.632142019271851,
|
|
"epoch": 0.9016811003565971,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.5451,
|
|
"mean_token_accuracy": 0.12307053580880165,
|
|
"num_tokens": 2083884.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.692302131652832,
|
|
"epoch": 0.9067753438614365,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.575,
|
|
"mean_token_accuracy": 0.11638538911938667,
|
|
"num_tokens": 2095429.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.6273736476898195,
|
|
"epoch": 0.9118695873662761,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5777,
|
|
"mean_token_accuracy": 0.11308116614818572,
|
|
"num_tokens": 2108351.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.797583866119385,
|
|
"epoch": 0.9169638308711157,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.6103,
|
|
"mean_token_accuracy": 0.11703968867659569,
|
|
"num_tokens": 2120227.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.566512250900269,
|
|
"epoch": 0.9220580743759552,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5417,
|
|
"mean_token_accuracy": 0.12239225804805756,
|
|
"num_tokens": 2131624.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.566377592086792,
|
|
"epoch": 0.9271523178807947,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5082,
|
|
"mean_token_accuracy": 0.11846095621585846,
|
|
"num_tokens": 2143523.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.590761947631836,
|
|
"epoch": 0.9322465613856342,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.12397759556770324,
|
|
"num_tokens": 2154952.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.632454824447632,
|
|
"epoch": 0.9373408048904738,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5431,
|
|
"mean_token_accuracy": 0.1184918574988842,
|
|
"num_tokens": 2166407.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.586945915222168,
|
|
"epoch": 0.9424350483953133,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5346,
|
|
"mean_token_accuracy": 0.11630669236183167,
|
|
"num_tokens": 2179748.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.59289584159851,
|
|
"epoch": 0.9475292919001528,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.617,
|
|
"mean_token_accuracy": 0.12137511149048805,
|
|
"num_tokens": 2191266.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.706100225448608,
|
|
"epoch": 0.9526235354049923,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.5404,
|
|
"mean_token_accuracy": 0.12512246817350386,
|
|
"num_tokens": 2203209.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.561947679519653,
|
|
"epoch": 0.9577177789098319,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5379,
|
|
"mean_token_accuracy": 0.12010324597358704,
|
|
"num_tokens": 2214693.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.549541568756103,
|
|
"epoch": 0.9628120224146715,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4383,
|
|
"mean_token_accuracy": 0.12597796693444252,
|
|
"num_tokens": 2226778.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.511363363265991,
|
|
"epoch": 0.9679062659195109,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5258,
|
|
"mean_token_accuracy": 0.11091761514544488,
|
|
"num_tokens": 2238528.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.628715085983276,
|
|
"epoch": 0.9730005094243505,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5948,
|
|
"mean_token_accuracy": 0.11641908511519432,
|
|
"num_tokens": 2251483.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.526347684860229,
|
|
"epoch": 0.97809475292919,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4457,
|
|
"mean_token_accuracy": 0.12605334371328353,
|
|
"num_tokens": 2262151.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.524391222000122,
|
|
"epoch": 0.9831889964340296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.12696175500750542,
|
|
"num_tokens": 2274533.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.529171466827393,
|
|
"epoch": 0.9882832399388691,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4701,
|
|
"mean_token_accuracy": 0.12050113677978516,
|
|
"num_tokens": 2285518.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.5718752384185795,
|
|
"epoch": 0.9933774834437086,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.5593,
|
|
"mean_token_accuracy": 0.1192559577524662,
|
|
"num_tokens": 2297900.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.551716423034668,
|
|
"epoch": 0.9984717269485481,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5026,
|
|
"mean_token_accuracy": 0.13176016956567765,
|
|
"num_tokens": 2311035.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.543602095709907,
|
|
"epoch": 1.0030565461029037,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3259,
|
|
"mean_token_accuracy": 0.13025057150257957,
|
|
"num_tokens": 2321287.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.562080955505371,
|
|
"epoch": 1.0081507896077433,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3717,
|
|
"mean_token_accuracy": 0.12037833482027054,
|
|
"num_tokens": 2333276.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.429567813873291,
|
|
"epoch": 1.0132450331125828,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.13079071342945098,
|
|
"num_tokens": 2346554.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.421297168731689,
|
|
"epoch": 1.0183392766174224,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.255,
|
|
"mean_token_accuracy": 0.13205485045909882,
|
|
"num_tokens": 2357564.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0183392766174224,
|
|
"eval_entropy": 6.351307462762903,
|
|
"eval_loss": 6.465834617614746,
|
|
"eval_mean_token_accuracy": 0.12646097975012696,
|
|
"eval_num_tokens": 2357564.0,
|
|
"eval_runtime": 2.7035,
|
|
"eval_samples_per_second": 1198.459,
|
|
"eval_steps_per_second": 149.807,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.417219495773315,
|
|
"epoch": 1.0234335201222617,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999997711134041,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.13421913608908653,
|
|
"num_tokens": 2369749.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.509257507324219,
|
|
"epoch": 1.0285277636271013,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999988412624068,
|
|
"loss": 6.3632,
|
|
"mean_token_accuracy": 0.12740975096821786,
|
|
"num_tokens": 2382622.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.493946552276611,
|
|
"epoch": 1.0336220071319409,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999971961445492,
|
|
"loss": 6.3463,
|
|
"mean_token_accuracy": 0.12851536944508551,
|
|
"num_tokens": 2394749.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.49571852684021,
|
|
"epoch": 1.0387162506367804,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999948357650613,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.12557510510087014,
|
|
"num_tokens": 2407868.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.418413162231445,
|
|
"epoch": 1.04381049414162,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999917601314468,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.1351904697716236,
|
|
"num_tokens": 2419045.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.423746490478516,
|
|
"epoch": 1.0489047376464595,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999879692534828,
|
|
"loss": 6.3565,
|
|
"mean_token_accuracy": 0.127122400701046,
|
|
"num_tokens": 2431697.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.501759767532349,
|
|
"epoch": 1.053998981151299,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999834631432208,
|
|
"loss": 6.3452,
|
|
"mean_token_accuracy": 0.12978592067956923,
|
|
"num_tokens": 2442579.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.348742437362671,
|
|
"epoch": 1.0590932246561386,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999782418149852,
|
|
"loss": 6.1903,
|
|
"mean_token_accuracy": 0.13898445516824723,
|
|
"num_tokens": 2453601.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.388893413543701,
|
|
"epoch": 1.064187468160978,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999723052853747,
|
|
"loss": 6.2305,
|
|
"mean_token_accuracy": 0.13225354850292206,
|
|
"num_tokens": 2465500.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.277715349197388,
|
|
"epoch": 1.0692817116658175,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999656535732615,
|
|
"loss": 6.256,
|
|
"mean_token_accuracy": 0.1289329580962658,
|
|
"num_tokens": 2477948.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.455788087844849,
|
|
"epoch": 1.074375955170657,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999582866997913,
|
|
"loss": 6.2824,
|
|
"mean_token_accuracy": 0.1348245456814766,
|
|
"num_tokens": 2489575.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.437306356430054,
|
|
"epoch": 1.0794701986754967,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499950204688383,
|
|
"loss": 6.2831,
|
|
"mean_token_accuracy": 0.1257219858467579,
|
|
"num_tokens": 2501615.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.318784904479981,
|
|
"epoch": 1.0845644421803362,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999414075647292,
|
|
"loss": 6.2665,
|
|
"mean_token_accuracy": 0.13291230499744416,
|
|
"num_tokens": 2513164.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.469882678985596,
|
|
"epoch": 1.0896586856851758,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000499931895356796,
|
|
"loss": 6.3728,
|
|
"mean_token_accuracy": 0.13190104067325592,
|
|
"num_tokens": 2524349.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.368467235565186,
|
|
"epoch": 1.0947529291900153,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999216680948223,
|
|
"loss": 6.2449,
|
|
"mean_token_accuracy": 0.12897544875741004,
|
|
"num_tokens": 2535834.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.293046236038208,
|
|
"epoch": 1.0998471726948549,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999107258113204,
|
|
"loss": 6.1801,
|
|
"mean_token_accuracy": 0.13650451526045798,
|
|
"num_tokens": 2547690.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.357415246963501,
|
|
"epoch": 1.1049414161996944,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998990685410759,
|
|
"loss": 6.2374,
|
|
"mean_token_accuracy": 0.13259905725717544,
|
|
"num_tokens": 2559249.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.391598272323608,
|
|
"epoch": 1.110035659704534,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998866963211465,
|
|
"loss": 6.3529,
|
|
"mean_token_accuracy": 0.1308108612895012,
|
|
"num_tokens": 2573014.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.377676296234131,
|
|
"epoch": 1.1151299032093733,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998736091908636,
|
|
"loss": 6.2418,
|
|
"mean_token_accuracy": 0.13509300872683525,
|
|
"num_tokens": 2584720.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.372010564804077,
|
|
"epoch": 1.1202241467142129,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998598071918309,
|
|
"loss": 6.2614,
|
|
"mean_token_accuracy": 0.1345648229122162,
|
|
"num_tokens": 2596711.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.382832479476929,
|
|
"epoch": 1.1253183902190524,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998452903679246,
|
|
"loss": 6.2082,
|
|
"mean_token_accuracy": 0.13323341384530069,
|
|
"num_tokens": 2606709.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.278462076187134,
|
|
"epoch": 1.130412633723892,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998300587652936,
|
|
"loss": 6.2504,
|
|
"mean_token_accuracy": 0.13396723121404647,
|
|
"num_tokens": 2617233.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.448177194595337,
|
|
"epoch": 1.1355068772287316,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998141124323586,
|
|
"loss": 6.2313,
|
|
"mean_token_accuracy": 0.13162466883659363,
|
|
"num_tokens": 2628313.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.1972558975219725,
|
|
"epoch": 1.1406011207335711,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997974514198126,
|
|
"loss": 6.1596,
|
|
"mean_token_accuracy": 0.1385033383965492,
|
|
"num_tokens": 2639198.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.381111478805542,
|
|
"epoch": 1.1456953642384107,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997800757806209,
|
|
"loss": 6.2595,
|
|
"mean_token_accuracy": 0.13240717723965645,
|
|
"num_tokens": 2650914.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.278826713562012,
|
|
"epoch": 1.15078960774325,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997619855700203,
|
|
"loss": 6.2428,
|
|
"mean_token_accuracy": 0.13494674637913703,
|
|
"num_tokens": 2664335.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.331632471084594,
|
|
"epoch": 1.1558838512480896,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499743180845519,
|
|
"loss": 6.3043,
|
|
"mean_token_accuracy": 0.1335383266210556,
|
|
"num_tokens": 2675839.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.197550296783447,
|
|
"epoch": 1.1609780947529291,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499723661666897,
|
|
"loss": 6.0912,
|
|
"mean_token_accuracy": 0.14703805819153787,
|
|
"num_tokens": 2686793.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.288230514526367,
|
|
"epoch": 1.1660723382577687,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997034280962054,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.12843396440148352,
|
|
"num_tokens": 2698504.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.3738954067230225,
|
|
"epoch": 1.1711665817626082,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996824801977663,
|
|
"loss": 6.1969,
|
|
"mean_token_accuracy": 0.13494786694645883,
|
|
"num_tokens": 2710678.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.318151998519897,
|
|
"epoch": 1.1762608252674478,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996608180381727,
|
|
"loss": 6.3212,
|
|
"mean_token_accuracy": 0.13294611647725105,
|
|
"num_tokens": 2722644.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.280697631835937,
|
|
"epoch": 1.1813550687722874,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499638441686288,
|
|
"loss": 6.1941,
|
|
"mean_token_accuracy": 0.13483970239758492,
|
|
"num_tokens": 2735279.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.307227325439453,
|
|
"epoch": 1.186449312277127,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996153512132462,
|
|
"loss": 6.2653,
|
|
"mean_token_accuracy": 0.13486257940530777,
|
|
"num_tokens": 2746905.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.3911340713500975,
|
|
"epoch": 1.1915435557819665,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995915466924517,
|
|
"loss": 6.3079,
|
|
"mean_token_accuracy": 0.13066884726285935,
|
|
"num_tokens": 2760081.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.279552936553955,
|
|
"epoch": 1.196637799286806,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995670281995784,
|
|
"loss": 6.2574,
|
|
"mean_token_accuracy": 0.13016019761562347,
|
|
"num_tokens": 2771416.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.394925212860107,
|
|
"epoch": 1.2017320427916454,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995417958125703,
|
|
"loss": 6.3563,
|
|
"mean_token_accuracy": 0.12249920591711998,
|
|
"num_tokens": 2784216.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.398853206634522,
|
|
"epoch": 1.206826286296485,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995158496116402,
|
|
"loss": 6.3365,
|
|
"mean_token_accuracy": 0.13531623110175134,
|
|
"num_tokens": 2795706.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.367800998687744,
|
|
"epoch": 1.2119205298013245,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004994891896792712,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.13469970226287842,
|
|
"num_tokens": 2806317.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.373945808410644,
|
|
"epoch": 1.217014773306164,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994618161002142,
|
|
"loss": 6.2667,
|
|
"mean_token_accuracy": 0.13018080592155457,
|
|
"num_tokens": 2817432.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.390775060653686,
|
|
"epoch": 1.2221090168110036,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994337289614893,
|
|
"loss": 6.2062,
|
|
"mean_token_accuracy": 0.13436981067061424,
|
|
"num_tokens": 2828145.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.194883871078491,
|
|
"epoch": 1.2272032603158431,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994049283523854,
|
|
"loss": 6.1799,
|
|
"mean_token_accuracy": 0.14733498990535737,
|
|
"num_tokens": 2838962.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.318769741058349,
|
|
"epoch": 1.2322975038206827,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004993754143644585,
|
|
"loss": 6.2335,
|
|
"mean_token_accuracy": 0.13520962446928025,
|
|
"num_tokens": 2850941.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.2552954196929935,
|
|
"epoch": 1.237391747325522,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004993451870915333,
|
|
"loss": 6.1947,
|
|
"mean_token_accuracy": 0.13884784355759622,
|
|
"num_tokens": 2863265.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.272042751312256,
|
|
"epoch": 1.2424859908303616,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004993142466297016,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.13500213846564293,
|
|
"num_tokens": 2876235.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.289383506774902,
|
|
"epoch": 1.2475802343352012,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004992825930773228,
|
|
"loss": 6.2629,
|
|
"mean_token_accuracy": 0.12987583950161935,
|
|
"num_tokens": 2889518.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.2832972526550295,
|
|
"epoch": 1.2526744778400407,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004992502265350224,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.13477277979254723,
|
|
"num_tokens": 2902417.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.272687149047852,
|
|
"epoch": 1.2577687213448803,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992171471056933,
|
|
"loss": 6.1611,
|
|
"mean_token_accuracy": 0.14579299539327623,
|
|
"num_tokens": 2912904.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.331518888473511,
|
|
"epoch": 1.2628629648497198,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004991833548944945,
|
|
"loss": 6.2294,
|
|
"mean_token_accuracy": 0.1335999645292759,
|
|
"num_tokens": 2925583.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.22144513130188,
|
|
"epoch": 1.2679572083545594,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991488500088505,
|
|
"loss": 6.1588,
|
|
"mean_token_accuracy": 0.14021274149417878,
|
|
"num_tokens": 2937916.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.189096450805664,
|
|
"epoch": 1.273051451859399,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004991136325584516,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.13783264458179473,
|
|
"num_tokens": 2949583.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.280501842498779,
|
|
"epoch": 1.2781456953642385,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004990777026552536,
|
|
"loss": 6.1986,
|
|
"mean_token_accuracy": 0.1347845807671547,
|
|
"num_tokens": 2962044.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.409678268432617,
|
|
"epoch": 1.283239938869078,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004990410604134769,
|
|
"loss": 6.3647,
|
|
"mean_token_accuracy": 0.12630252316594123,
|
|
"num_tokens": 2973570.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.199240875244141,
|
|
"epoch": 1.2883341823739174,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004990037059496064,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.13618834540247918,
|
|
"num_tokens": 2983683.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.397557020187378,
|
|
"epoch": 1.293428425878757,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989656393823913,
|
|
"loss": 6.2489,
|
|
"mean_token_accuracy": 0.13301505297422409,
|
|
"num_tokens": 2995100.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.195611143112183,
|
|
"epoch": 1.2985226693835965,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004989268608328443,
|
|
"loss": 6.2716,
|
|
"mean_token_accuracy": 0.1302342250943184,
|
|
"num_tokens": 3006887.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.336117124557495,
|
|
"epoch": 1.303616912888436,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004988873704242418,
|
|
"loss": 6.2255,
|
|
"mean_token_accuracy": 0.13720177486538887,
|
|
"num_tokens": 3020444.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.2508533000946045,
|
|
"epoch": 1.3087111563932756,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004988471682821229,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.13537554293870926,
|
|
"num_tokens": 3032487.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.28198561668396,
|
|
"epoch": 1.3138053998981152,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004988062545342894,
|
|
"loss": 6.2708,
|
|
"mean_token_accuracy": 0.13203708529472352,
|
|
"num_tokens": 3044652.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.253536605834961,
|
|
"epoch": 1.3188996434029547,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004987646293108056,
|
|
"loss": 6.1579,
|
|
"mean_token_accuracy": 0.13946222364902497,
|
|
"num_tokens": 3056252.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.187241125106811,
|
|
"epoch": 1.323993886907794,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004987222927439967,
|
|
"loss": 6.175,
|
|
"mean_token_accuracy": 0.14278096705675125,
|
|
"num_tokens": 3066959.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.314873886108399,
|
|
"epoch": 1.3290881304126336,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986792449684504,
|
|
"loss": 6.2857,
|
|
"mean_token_accuracy": 0.13692365884780883,
|
|
"num_tokens": 3079113.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.333209562301636,
|
|
"epoch": 1.3341823739174732,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986354861210144,
|
|
"loss": 6.2074,
|
|
"mean_token_accuracy": 0.13738809376955033,
|
|
"num_tokens": 3090290.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.23097414970398,
|
|
"epoch": 1.3392766174223127,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004985910163407973,
|
|
"loss": 6.1646,
|
|
"mean_token_accuracy": 0.14096444249153137,
|
|
"num_tokens": 3101031.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.222257280349732,
|
|
"epoch": 1.3443708609271523,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004985458357691675,
|
|
"loss": 6.2189,
|
|
"mean_token_accuracy": 0.13166088685393335,
|
|
"num_tokens": 3113371.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.228028869628906,
|
|
"epoch": 1.3494651044319919,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004984999445497533,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.13785334452986717,
|
|
"num_tokens": 3125167.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.183880138397217,
|
|
"epoch": 1.3545593479368314,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000498453342828442,
|
|
"loss": 6.24,
|
|
"mean_token_accuracy": 0.1362723782658577,
|
|
"num_tokens": 3136945.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.322224140167236,
|
|
"epoch": 1.359653591441671,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984060307533794,
|
|
"loss": 6.1555,
|
|
"mean_token_accuracy": 0.14208487123250962,
|
|
"num_tokens": 3147435.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.202198314666748,
|
|
"epoch": 1.3647478349465105,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004983580084749698,
|
|
"loss": 6.1869,
|
|
"mean_token_accuracy": 0.1455744594335556,
|
|
"num_tokens": 3157693.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.214845132827759,
|
|
"epoch": 1.36984207845135,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000498309276145875,
|
|
"loss": 6.1204,
|
|
"mean_token_accuracy": 0.14093155562877654,
|
|
"num_tokens": 3169561.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.275221824645996,
|
|
"epoch": 1.3749363219561896,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004982598339210142,
|
|
"loss": 6.3096,
|
|
"mean_token_accuracy": 0.12878431007266045,
|
|
"num_tokens": 3183142.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.235415983200073,
|
|
"epoch": 1.380030565461029,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004982096819575631,
|
|
"loss": 6.2222,
|
|
"mean_token_accuracy": 0.13532169610261918,
|
|
"num_tokens": 3195703.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.144252157211303,
|
|
"epoch": 1.3851248089658685,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004981588204149538,
|
|
"loss": 6.1176,
|
|
"mean_token_accuracy": 0.14307456761598586,
|
|
"num_tokens": 3206603.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.285677289962768,
|
|
"epoch": 1.390219052470708,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004981072494548743,
|
|
"loss": 6.2249,
|
|
"mean_token_accuracy": 0.13540513813495636,
|
|
"num_tokens": 3218909.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.245230484008789,
|
|
"epoch": 1.3953132959755477,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004980549692412676,
|
|
"loss": 6.1677,
|
|
"mean_token_accuracy": 0.1433960661292076,
|
|
"num_tokens": 3230252.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.1964904308319095,
|
|
"epoch": 1.4004075394803872,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004980019799403313,
|
|
"loss": 6.2193,
|
|
"mean_token_accuracy": 0.13967160061001777,
|
|
"num_tokens": 3241227.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.211932706832886,
|
|
"epoch": 1.4055017829852268,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004979482817205176,
|
|
"loss": 6.0619,
|
|
"mean_token_accuracy": 0.14620037674903869,
|
|
"num_tokens": 3253569.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.2361126899719235,
|
|
"epoch": 1.410596026490066,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004978938747525319,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.138123869150877,
|
|
"num_tokens": 3265371.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.251908636093139,
|
|
"epoch": 1.4156902699949057,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004978387592093329,
|
|
"loss": 6.2013,
|
|
"mean_token_accuracy": 0.13564864099025725,
|
|
"num_tokens": 3276612.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.239566612243652,
|
|
"epoch": 1.4207845134997452,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004977829352661318,
|
|
"loss": 6.2196,
|
|
"mean_token_accuracy": 0.1351855956017971,
|
|
"num_tokens": 3288847.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.063872957229615,
|
|
"epoch": 1.4258787570045848,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004977264031003919,
|
|
"loss": 6.1702,
|
|
"mean_token_accuracy": 0.14193566218018533,
|
|
"num_tokens": 3300738.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.229835319519043,
|
|
"epoch": 1.4309730005094243,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004976691628918278,
|
|
"loss": 6.1568,
|
|
"mean_token_accuracy": 0.14138704612851144,
|
|
"num_tokens": 3312369.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.212840414047241,
|
|
"epoch": 1.4360672440142639,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000497611214822405,
|
|
"loss": 6.1656,
|
|
"mean_token_accuracy": 0.1398153379559517,
|
|
"num_tokens": 3325277.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.102501726150512,
|
|
"epoch": 1.4411614875191034,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004975525590763393,
|
|
"loss": 6.0414,
|
|
"mean_token_accuracy": 0.1459734968841076,
|
|
"num_tokens": 3335881.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.301107358932495,
|
|
"epoch": 1.446255731023943,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000497493195840096,
|
|
"loss": 6.2092,
|
|
"mean_token_accuracy": 0.13697200790047645,
|
|
"num_tokens": 3348942.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.175320100784302,
|
|
"epoch": 1.4513499745287826,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004974331253023898,
|
|
"loss": 6.2037,
|
|
"mean_token_accuracy": 0.14206972047686578,
|
|
"num_tokens": 3361890.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.227373933792114,
|
|
"epoch": 1.4564442180336221,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004973723476541836,
|
|
"loss": 6.179,
|
|
"mean_token_accuracy": 0.1412748746573925,
|
|
"num_tokens": 3373297.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.052916145324707,
|
|
"epoch": 1.4615384615384617,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004973108630886885,
|
|
"loss": 6.0592,
|
|
"mean_token_accuracy": 0.14644164815545083,
|
|
"num_tokens": 3384623.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.1409101486206055,
|
|
"epoch": 1.466632705043301,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004972486718013626,
|
|
"loss": 6.1425,
|
|
"mean_token_accuracy": 0.14146411642432213,
|
|
"num_tokens": 3398379.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.276405096054077,
|
|
"epoch": 1.4717269485481406,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004971857739899109,
|
|
"loss": 6.1401,
|
|
"mean_token_accuracy": 0.13381285667419435,
|
|
"num_tokens": 3410928.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.092034387588501,
|
|
"epoch": 1.4768211920529801,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004971221698542842,
|
|
"loss": 6.0856,
|
|
"mean_token_accuracy": 0.1479926437139511,
|
|
"num_tokens": 3422173.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.1313581466674805,
|
|
"epoch": 1.4819154355578197,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004970578595966789,
|
|
"loss": 6.1405,
|
|
"mean_token_accuracy": 0.14326228201389313,
|
|
"num_tokens": 3434341.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.1404660701751705,
|
|
"epoch": 1.4870096790626592,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004969928434215358,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.13853721991181372,
|
|
"num_tokens": 3445605.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.229519939422607,
|
|
"epoch": 1.4921039225674988,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004969271215355402,
|
|
"loss": 6.1073,
|
|
"mean_token_accuracy": 0.14102545380592346,
|
|
"num_tokens": 3456957.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.172272682189941,
|
|
"epoch": 1.4971981660723381,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004968606941476207,
|
|
"loss": 6.1453,
|
|
"mean_token_accuracy": 0.14370948895812036,
|
|
"num_tokens": 3468340.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.181661653518677,
|
|
"epoch": 1.5022924095771777,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004967935614689485,
|
|
"loss": 6.1106,
|
|
"mean_token_accuracy": 0.14410856515169143,
|
|
"num_tokens": 3479137.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.121802949905396,
|
|
"epoch": 1.5073866530820172,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004967257237129373,
|
|
"loss": 6.0129,
|
|
"mean_token_accuracy": 0.15014424845576285,
|
|
"num_tokens": 3490525.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.079214286804199,
|
|
"epoch": 1.5124808965868568,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004966571810952417,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.15413511991500856,
|
|
"num_tokens": 3500731.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.110242938995361,
|
|
"epoch": 1.5175751400916964,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004965879338337577,
|
|
"loss": 6.1093,
|
|
"mean_token_accuracy": 0.14118633270263672,
|
|
"num_tokens": 3512944.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.12162446975708,
|
|
"epoch": 1.522669383596536,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004965179821486205,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.13991874754428862,
|
|
"num_tokens": 3524110.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.107229280471802,
|
|
"epoch": 1.5277636271013755,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004964473262622054,
|
|
"loss": 6.1368,
|
|
"mean_token_accuracy": 0.14842215329408645,
|
|
"num_tokens": 3536162.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5277636271013755,
|
|
"eval_entropy": 6.009506080768727,
|
|
"eval_loss": 6.184627056121826,
|
|
"eval_mean_token_accuracy": 0.14316514546488537,
|
|
"eval_num_tokens": 3536162.0,
|
|
"eval_runtime": 2.6604,
|
|
"eval_samples_per_second": 1217.879,
|
|
"eval_steps_per_second": 152.235,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.133478784561158,
|
|
"epoch": 1.532857870606215,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496375966399126,
|
|
"loss": 6.0538,
|
|
"mean_token_accuracy": 0.1511278361082077,
|
|
"num_tokens": 3547756.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.997680425643921,
|
|
"epoch": 1.5379521141110546,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000496303902786234,
|
|
"loss": 6.0479,
|
|
"mean_token_accuracy": 0.15588610172271727,
|
|
"num_tokens": 3559218.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.21874737739563,
|
|
"epoch": 1.5430463576158941,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004962311356526181,
|
|
"loss": 6.1004,
|
|
"mean_token_accuracy": 0.14346015974879264,
|
|
"num_tokens": 3571875.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.06115870475769,
|
|
"epoch": 1.5481406011207337,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004961576652296036,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.14653350710868834,
|
|
"num_tokens": 3583297.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.078774929046631,
|
|
"epoch": 1.5532348446255733,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004960834917507516,
|
|
"loss": 6.0777,
|
|
"mean_token_accuracy": 0.14559792429208757,
|
|
"num_tokens": 3595302.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.114213848114014,
|
|
"epoch": 1.5583290881304126,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004960086154518578,
|
|
"loss": 6.1155,
|
|
"mean_token_accuracy": 0.1444586381316185,
|
|
"num_tokens": 3606940.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.140858173370361,
|
|
"epoch": 1.5634233316352522,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004959330365709528,
|
|
"loss": 6.1742,
|
|
"mean_token_accuracy": 0.13936522379517555,
|
|
"num_tokens": 3619364.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.136679077148438,
|
|
"epoch": 1.5685175751400917,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004958567553483003,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.1398473523557186,
|
|
"num_tokens": 3631789.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.210715007781983,
|
|
"epoch": 1.5736118186449313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004957797720263967,
|
|
"loss": 6.0968,
|
|
"mean_token_accuracy": 0.1433316245675087,
|
|
"num_tokens": 3643126.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.055673503875733,
|
|
"epoch": 1.5787060621497706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004957020868499706,
|
|
"loss": 6.1212,
|
|
"mean_token_accuracy": 0.14359405040740966,
|
|
"num_tokens": 3654253.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.220635080337525,
|
|
"epoch": 1.5838003056546102,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004956237000659815,
|
|
"loss": 6.0909,
|
|
"mean_token_accuracy": 0.14231853038072587,
|
|
"num_tokens": 3666025.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.071724557876587,
|
|
"epoch": 1.5888945491594497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004955446119236195,
|
|
"loss": 6.1059,
|
|
"mean_token_accuracy": 0.14634145349264144,
|
|
"num_tokens": 3676103.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.272754764556884,
|
|
"epoch": 1.5939887926642893,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004954648226743043,
|
|
"loss": 6.2349,
|
|
"mean_token_accuracy": 0.13702599853277206,
|
|
"num_tokens": 3688956.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.046952247619629,
|
|
"epoch": 1.5990830361691288,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004953843325716841,
|
|
"loss": 6.0908,
|
|
"mean_token_accuracy": 0.1434723451733589,
|
|
"num_tokens": 3700288.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.251407861709595,
|
|
"epoch": 1.6041772796739684,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004953031418716356,
|
|
"loss": 6.1313,
|
|
"mean_token_accuracy": 0.1386256992816925,
|
|
"num_tokens": 3711342.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.078168249130249,
|
|
"epoch": 1.609271523178808,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004952212508322623,
|
|
"loss": 6.0284,
|
|
"mean_token_accuracy": 0.15199928134679794,
|
|
"num_tokens": 3722195.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.161301946640014,
|
|
"epoch": 1.6143657666836475,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004951386597138943,
|
|
"loss": 6.0596,
|
|
"mean_token_accuracy": 0.13910518661141397,
|
|
"num_tokens": 3733345.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.0794189453125,
|
|
"epoch": 1.619460010188487,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004950553687790871,
|
|
"loss": 6.1234,
|
|
"mean_token_accuracy": 0.14309294521808624,
|
|
"num_tokens": 3746975.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.060288143157959,
|
|
"epoch": 1.6245542536933266,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004949713782926208,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.14610447585582734,
|
|
"num_tokens": 3758027.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.145478296279907,
|
|
"epoch": 1.6296484971981662,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004948866885214996,
|
|
"loss": 6.024,
|
|
"mean_token_accuracy": 0.1490251824259758,
|
|
"num_tokens": 3769046.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.070356655120849,
|
|
"epoch": 1.6347427407030057,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004948012997349506,
|
|
"loss": 6.1225,
|
|
"mean_token_accuracy": 0.13770004659891127,
|
|
"num_tokens": 3781487.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.106816005706787,
|
|
"epoch": 1.6398369842078453,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004947152122044232,
|
|
"loss": 6.0784,
|
|
"mean_token_accuracy": 0.1434733733534813,
|
|
"num_tokens": 3793089.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.085491609573364,
|
|
"epoch": 1.6449312277126846,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004946284262035875,
|
|
"loss": 6.0534,
|
|
"mean_token_accuracy": 0.1486929565668106,
|
|
"num_tokens": 3804871.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.07642707824707,
|
|
"epoch": 1.6500254712175242,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494540942008335,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.1559026300907135,
|
|
"num_tokens": 3815258.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.128446197509765,
|
|
"epoch": 1.6551197147223637,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000494452759896776,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.14560223817825318,
|
|
"num_tokens": 3826578.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.194732284545898,
|
|
"epoch": 1.6602139582272033,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004943638801492396,
|
|
"loss": 6.0686,
|
|
"mean_token_accuracy": 0.14353293254971505,
|
|
"num_tokens": 3837675.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.103059434890747,
|
|
"epoch": 1.6653082017320426,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004942743030482728,
|
|
"loss": 6.0959,
|
|
"mean_token_accuracy": 0.14127544909715653,
|
|
"num_tokens": 3850108.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.089843034744263,
|
|
"epoch": 1.6704024452368822,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004941840288786397,
|
|
"loss": 5.9985,
|
|
"mean_token_accuracy": 0.14871155768632888,
|
|
"num_tokens": 3861975.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.037532472610474,
|
|
"epoch": 1.6754966887417218,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004940930579273198,
|
|
"loss": 6.0814,
|
|
"mean_token_accuracy": 0.14704819321632384,
|
|
"num_tokens": 3873819.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.140864372253418,
|
|
"epoch": 1.6805909322465613,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004940013904835079,
|
|
"loss": 6.1276,
|
|
"mean_token_accuracy": 0.14329199567437173,
|
|
"num_tokens": 3885821.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.020464944839477,
|
|
"epoch": 1.6856851757514009,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004939090268386132,
|
|
"loss": 6.061,
|
|
"mean_token_accuracy": 0.14725224077701568,
|
|
"num_tokens": 3897078.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.177228784561157,
|
|
"epoch": 1.6907794192562404,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004938159672862578,
|
|
"loss": 6.1186,
|
|
"mean_token_accuracy": 0.14301334992051123,
|
|
"num_tokens": 3909073.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.1155922412872314,
|
|
"epoch": 1.69587366276108,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004937222121222763,
|
|
"loss": 6.0403,
|
|
"mean_token_accuracy": 0.15009486079216003,
|
|
"num_tokens": 3920209.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.1138163089752195,
|
|
"epoch": 1.7009679062659195,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004936277616447144,
|
|
"loss": 6.0812,
|
|
"mean_token_accuracy": 0.14103334173560142,
|
|
"num_tokens": 3933202.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.031678533554077,
|
|
"epoch": 1.706062149770759,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004935326161538281,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.15366721600294114,
|
|
"num_tokens": 3944574.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.11631498336792,
|
|
"epoch": 1.7111563932755987,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004934367759520834,
|
|
"loss": 6.1349,
|
|
"mean_token_accuracy": 0.13778666779398918,
|
|
"num_tokens": 3956614.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.039500665664673,
|
|
"epoch": 1.7162506367804382,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004933402413441541,
|
|
"loss": 5.9878,
|
|
"mean_token_accuracy": 0.14830894619226456,
|
|
"num_tokens": 3967854.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.038022136688232,
|
|
"epoch": 1.7213448802852778,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004932430126369221,
|
|
"loss": 6.0042,
|
|
"mean_token_accuracy": 0.15413039326667785,
|
|
"num_tokens": 3980267.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.1142689228057865,
|
|
"epoch": 1.7264391237901173,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004931450901394753,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.14080657660961152,
|
|
"num_tokens": 3991551.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.961064291000366,
|
|
"epoch": 1.7315333672949567,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004930464741631073,
|
|
"loss": 5.9363,
|
|
"mean_token_accuracy": 0.15858148485422136,
|
|
"num_tokens": 4001685.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.097273063659668,
|
|
"epoch": 1.7366276107997962,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004929471650213167,
|
|
"loss": 6.023,
|
|
"mean_token_accuracy": 0.1522778809070587,
|
|
"num_tokens": 4011960.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.09654483795166,
|
|
"epoch": 1.7417218543046358,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004928471630298051,
|
|
"loss": 6.0452,
|
|
"mean_token_accuracy": 0.14763895571231841,
|
|
"num_tokens": 4023608.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.096526288986206,
|
|
"epoch": 1.7468160978094753,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004927464685064767,
|
|
"loss": 6.0664,
|
|
"mean_token_accuracy": 0.14474863559007645,
|
|
"num_tokens": 4035365.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.047125577926636,
|
|
"epoch": 1.7519103413143147,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004926450817714376,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.1497434765100479,
|
|
"num_tokens": 4046796.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.0006410598754885,
|
|
"epoch": 1.7570045848191542,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004925430031469941,
|
|
"loss": 6.0487,
|
|
"mean_token_accuracy": 0.13952309116721154,
|
|
"num_tokens": 4057749.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.124393129348755,
|
|
"epoch": 1.7620988283239938,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004924402329576522,
|
|
"loss": 6.1128,
|
|
"mean_token_accuracy": 0.14142398536205292,
|
|
"num_tokens": 4069491.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.0156410217285154,
|
|
"epoch": 1.7671930718288333,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004923367715301164,
|
|
"loss": 5.9604,
|
|
"mean_token_accuracy": 0.149125587195158,
|
|
"num_tokens": 4081552.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.112949228286743,
|
|
"epoch": 1.772287315333673,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004922326191932882,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.14075233936309814,
|
|
"num_tokens": 4094701.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.994672060012817,
|
|
"epoch": 1.7773815588385125,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004921277762782661,
|
|
"loss": 5.9122,
|
|
"mean_token_accuracy": 0.15512796491384506,
|
|
"num_tokens": 4105202.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.082731771469116,
|
|
"epoch": 1.782475802343352,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004920222431183437,
|
|
"loss": 6.1068,
|
|
"mean_token_accuracy": 0.14282587841153144,
|
|
"num_tokens": 4116447.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.129996824264526,
|
|
"epoch": 1.7875700458481916,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004919160200490085,
|
|
"loss": 6.1383,
|
|
"mean_token_accuracy": 0.14118616431951522,
|
|
"num_tokens": 4128103.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.121463251113892,
|
|
"epoch": 1.7926642893530311,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004918091074079419,
|
|
"loss": 6.1258,
|
|
"mean_token_accuracy": 0.14120445325970649,
|
|
"num_tokens": 4140259.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.058639049530029,
|
|
"epoch": 1.7977585328578707,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004917015055350167,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.15162411332130432,
|
|
"num_tokens": 4151513.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.050025558471679,
|
|
"epoch": 1.8028527763627102,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004915932147722974,
|
|
"loss": 6.0446,
|
|
"mean_token_accuracy": 0.1424715355038643,
|
|
"num_tokens": 4163539.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.072246265411377,
|
|
"epoch": 1.8079470198675498,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004914842354640382,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.14208936542272568,
|
|
"num_tokens": 4177173.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.0362903594970705,
|
|
"epoch": 1.8130412633723894,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000491374567956682,
|
|
"loss": 6.0433,
|
|
"mean_token_accuracy": 0.1477975770831108,
|
|
"num_tokens": 4189756.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.149643087387085,
|
|
"epoch": 1.8181355068772287,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004912642125988597,
|
|
"loss": 6.1003,
|
|
"mean_token_accuracy": 0.1537694126367569,
|
|
"num_tokens": 4203594.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.061221504211426,
|
|
"epoch": 1.8232297503820682,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004911531697413888,
|
|
"loss": 6.0252,
|
|
"mean_token_accuracy": 0.1486793577671051,
|
|
"num_tokens": 4216189.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.0181516170501705,
|
|
"epoch": 1.8283239938869078,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004910414397372724,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.1441796153783798,
|
|
"num_tokens": 4228971.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.129302215576172,
|
|
"epoch": 1.8334182373917474,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004909290229416979,
|
|
"loss": 6.0547,
|
|
"mean_token_accuracy": 0.1431230567395687,
|
|
"num_tokens": 4240671.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.030335569381714,
|
|
"epoch": 1.8385124808965867,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000490815919712036,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.1441195696592331,
|
|
"num_tokens": 4253522.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.055653190612793,
|
|
"epoch": 1.8436067244014263,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004907021304078398,
|
|
"loss": 6.0676,
|
|
"mean_token_accuracy": 0.1464722380042076,
|
|
"num_tokens": 4266545.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.024935054779053,
|
|
"epoch": 1.8487009679062658,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004905876553908432,
|
|
"loss": 5.9734,
|
|
"mean_token_accuracy": 0.15227250307798385,
|
|
"num_tokens": 4278692.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.068737316131592,
|
|
"epoch": 1.8537952114111054,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004904724950249599,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.15321767032146455,
|
|
"num_tokens": 4291154.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.084896469116211,
|
|
"epoch": 1.858889454915945,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004903566496762824,
|
|
"loss": 6.0975,
|
|
"mean_token_accuracy": 0.14444440975785255,
|
|
"num_tokens": 4303963.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.955971765518188,
|
|
"epoch": 1.8639836984207845,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000490240119713081,
|
|
"loss": 5.894,
|
|
"mean_token_accuracy": 0.15113551169633865,
|
|
"num_tokens": 4316046.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.058926010131836,
|
|
"epoch": 1.869077941925624,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004901229055058019,
|
|
"loss": 5.9533,
|
|
"mean_token_accuracy": 0.15235763043165207,
|
|
"num_tokens": 4327631.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.0447791576385494,
|
|
"epoch": 1.8741721854304636,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004900050074270669,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.15083706974983216,
|
|
"num_tokens": 4338994.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.9091675758361815,
|
|
"epoch": 1.8792664289353032,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004898864258516717,
|
|
"loss": 5.9759,
|
|
"mean_token_accuracy": 0.158118836581707,
|
|
"num_tokens": 4351197.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.0877978801727295,
|
|
"epoch": 1.8843606724401427,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004897671611565845,
|
|
"loss": 6.0025,
|
|
"mean_token_accuracy": 0.14755566269159318,
|
|
"num_tokens": 4363497.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.058438396453857,
|
|
"epoch": 1.8894549159449823,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004896472137209457,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.14204568788409233,
|
|
"num_tokens": 4376518.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.988859415054321,
|
|
"epoch": 1.8945491594498218,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004895265839260657,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.1513100005686283,
|
|
"num_tokens": 4387905.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.94207067489624,
|
|
"epoch": 1.8996434029546614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004894052721554242,
|
|
"loss": 5.9952,
|
|
"mean_token_accuracy": 0.15157082080841064,
|
|
"num_tokens": 4400188.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.07070140838623,
|
|
"epoch": 1.9047376464595007,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004892832787946691,
|
|
"loss": 6.0288,
|
|
"mean_token_accuracy": 0.144869889318943,
|
|
"num_tokens": 4411578.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.027485084533692,
|
|
"epoch": 1.9098318899643403,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004891606042316146,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.1512698881328106,
|
|
"num_tokens": 4423405.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.933491897583008,
|
|
"epoch": 1.9149261334691798,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004890372488562408,
|
|
"loss": 5.9792,
|
|
"mean_token_accuracy": 0.1508818507194519,
|
|
"num_tokens": 4435049.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.040270662307739,
|
|
"epoch": 1.9200203769740194,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004889132130606921,
|
|
"loss": 5.9578,
|
|
"mean_token_accuracy": 0.15416083931922914,
|
|
"num_tokens": 4445613.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.975605201721192,
|
|
"epoch": 1.9251146204788587,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004887884972392758,
|
|
"loss": 5.9572,
|
|
"mean_token_accuracy": 0.1484675571322441,
|
|
"num_tokens": 4458356.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.027165937423706,
|
|
"epoch": 1.9302088639836983,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000488663101788461,
|
|
"loss": 5.9582,
|
|
"mean_token_accuracy": 0.14620108753442765,
|
|
"num_tokens": 4470279.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.105115556716919,
|
|
"epoch": 1.9353031074885378,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004885370271068773,
|
|
"loss": 6.1719,
|
|
"mean_token_accuracy": 0.14128794223070146,
|
|
"num_tokens": 4481611.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.095979356765747,
|
|
"epoch": 1.9403973509933774,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004884102735953136,
|
|
"loss": 5.9509,
|
|
"mean_token_accuracy": 0.15084906220436095,
|
|
"num_tokens": 4493384.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.874639368057251,
|
|
"epoch": 1.945491594498217,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000488282841656717,
|
|
"loss": 5.8558,
|
|
"mean_token_accuracy": 0.15816857516765595,
|
|
"num_tokens": 4504059.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.0805902004241945,
|
|
"epoch": 1.9505858380030565,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048815473169619097,
|
|
"loss": 6.0322,
|
|
"mean_token_accuracy": 0.14308103621006013,
|
|
"num_tokens": 4516155.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.023840379714966,
|
|
"epoch": 1.955680081507896,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004880259441209945,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.15768046155571938,
|
|
"num_tokens": 4528030.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.012329912185669,
|
|
"epoch": 1.9607743250127356,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004878964793405408,
|
|
"loss": 6.0323,
|
|
"mean_token_accuracy": 0.1385513499379158,
|
|
"num_tokens": 4539217.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.005829763412476,
|
|
"epoch": 1.9658685685175752,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048776633776639584,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.159397454559803,
|
|
"num_tokens": 4550524.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.987122201919556,
|
|
"epoch": 1.9709628120224147,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048763551981227703,
|
|
"loss": 5.9538,
|
|
"mean_token_accuracy": 0.15249492228031158,
|
|
"num_tokens": 4561693.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.944087219238281,
|
|
"epoch": 1.9760570555272543,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048750402589405224,
|
|
"loss": 5.9474,
|
|
"mean_token_accuracy": 0.14776530414819716,
|
|
"num_tokens": 4573971.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.08344521522522,
|
|
"epoch": 1.9811512990320939,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048737185642973784,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.15248930752277373,
|
|
"num_tokens": 4584870.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.944841814041138,
|
|
"epoch": 1.9862455425369334,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000487239011839498,
|
|
"loss": 6.0011,
|
|
"mean_token_accuracy": 0.1480586051940918,
|
|
"num_tokens": 4595973.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.001732110977173,
|
|
"epoch": 1.9913397860417728,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004871054925456431,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.15668184459209442,
|
|
"num_tokens": 4607516.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.914603996276855,
|
|
"epoch": 1.9964340295466123,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048697129897262833,
|
|
"loss": 5.9035,
|
|
"mean_token_accuracy": 0.1592019245028496,
|
|
"num_tokens": 4619382.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.0304126209682885,
|
|
"epoch": 2.001018848700968,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004868364315470523,
|
|
"loss": 5.9977,
|
|
"mean_token_accuracy": 0.1434525102376938,
|
|
"num_tokens": 4630049.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.960834646224976,
|
|
"epoch": 2.0061130922058075,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048670089069765594,
|
|
"loss": 5.6245,
|
|
"mean_token_accuracy": 0.16735252141952514,
|
|
"num_tokens": 4642919.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.999763917922974,
|
|
"epoch": 2.011207335710647,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004865646768553209,
|
|
"loss": 5.7162,
|
|
"mean_token_accuracy": 0.1654820777475834,
|
|
"num_tokens": 4656620.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.935464525222779,
|
|
"epoch": 2.0163015792154866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004864277904530684,
|
|
"loss": 5.6283,
|
|
"mean_token_accuracy": 0.1659976363182068,
|
|
"num_tokens": 4668764.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.838066720962525,
|
|
"epoch": 2.021395822720326,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048629023192605747,
|
|
"loss": 5.5223,
|
|
"mean_token_accuracy": 0.17317576110363006,
|
|
"num_tokens": 4680014.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.937760353088379,
|
|
"epoch": 2.0264900662251657,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048615200171158395,
|
|
"loss": 5.5979,
|
|
"mean_token_accuracy": 0.1677281975746155,
|
|
"num_tokens": 4692677.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.756949186325073,
|
|
"epoch": 2.0315843097300053,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048601310024907904,
|
|
"loss": 5.5487,
|
|
"mean_token_accuracy": 0.17209948599338531,
|
|
"num_tokens": 4704585.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.923199081420899,
|
|
"epoch": 2.036678553234845,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048587352798010766,
|
|
"loss": 5.6052,
|
|
"mean_token_accuracy": 0.16866799145936967,
|
|
"num_tokens": 4716418.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.036678553234845,
|
|
"eval_entropy": 5.7684895903975875,
|
|
"eval_loss": 5.99842643737793,
|
|
"eval_mean_token_accuracy": 0.15436349115621897,
|
|
"eval_num_tokens": 4716418.0,
|
|
"eval_runtime": 2.6838,
|
|
"eval_samples_per_second": 1207.227,
|
|
"eval_steps_per_second": 150.903,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9810,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 7536964580352000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|