2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0183392766174224,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3752578179072000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|