lllqaq's picture
Upload checkpoint-800 (full checkpoint incl. ZeRO states)
50137dc verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.9801980198019802,
"eval_steps": 500,
"global_step": 800,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.024752475247524754,
"grad_norm": 2.377986153981849,
"learning_rate": 1.111111111111111e-06,
"loss": 0.5134,
"step": 10
},
{
"epoch": 0.04950495049504951,
"grad_norm": 1.1527847734451684,
"learning_rate": 2.345679012345679e-06,
"loss": 0.4478,
"step": 20
},
{
"epoch": 0.07425742574257425,
"grad_norm": 0.9214168013745976,
"learning_rate": 3.580246913580247e-06,
"loss": 0.3606,
"step": 30
},
{
"epoch": 0.09900990099009901,
"grad_norm": 0.6169142314211664,
"learning_rate": 4.814814814814815e-06,
"loss": 0.2914,
"step": 40
},
{
"epoch": 0.12376237623762376,
"grad_norm": 0.9045543763732307,
"learning_rate": 6.049382716049383e-06,
"loss": 0.258,
"step": 50
},
{
"epoch": 0.1485148514851485,
"grad_norm": 0.6916239240771108,
"learning_rate": 7.283950617283952e-06,
"loss": 0.2541,
"step": 60
},
{
"epoch": 0.17326732673267325,
"grad_norm": 0.6563868068913417,
"learning_rate": 8.518518518518519e-06,
"loss": 0.2375,
"step": 70
},
{
"epoch": 0.19801980198019803,
"grad_norm": 0.6955688922095051,
"learning_rate": 9.753086419753087e-06,
"loss": 0.2281,
"step": 80
},
{
"epoch": 0.22277227722772278,
"grad_norm": 0.7774430403471241,
"learning_rate": 9.997012501794273e-06,
"loss": 0.2409,
"step": 90
},
{
"epoch": 0.24752475247524752,
"grad_norm": 0.7176934124930091,
"learning_rate": 9.984881908680157e-06,
"loss": 0.2235,
"step": 100
},
{
"epoch": 0.2722772277227723,
"grad_norm": 0.6385467956457365,
"learning_rate": 9.963444133394478e-06,
"loss": 0.2309,
"step": 110
},
{
"epoch": 0.297029702970297,
"grad_norm": 0.7973847481250326,
"learning_rate": 9.93273920201681e-06,
"loss": 0.2123,
"step": 120
},
{
"epoch": 0.3217821782178218,
"grad_norm": 0.7212534759324712,
"learning_rate": 9.892824443164987e-06,
"loss": 0.2261,
"step": 130
},
{
"epoch": 0.3465346534653465,
"grad_norm": 0.7370462388577308,
"learning_rate": 9.84377438095789e-06,
"loss": 0.2256,
"step": 140
},
{
"epoch": 0.3712871287128713,
"grad_norm": 0.7271303244318778,
"learning_rate": 9.785680595872824e-06,
"loss": 0.2241,
"step": 150
},
{
"epoch": 0.39603960396039606,
"grad_norm": 0.5777488399063054,
"learning_rate": 9.718651553757266e-06,
"loss": 0.2093,
"step": 160
},
{
"epoch": 0.4207920792079208,
"grad_norm": 0.6578719601127735,
"learning_rate": 9.642812403314272e-06,
"loss": 0.2168,
"step": 170
},
{
"epoch": 0.44554455445544555,
"grad_norm": 0.6851661976650275,
"learning_rate": 9.55830474243961e-06,
"loss": 0.2181,
"step": 180
},
{
"epoch": 0.47029702970297027,
"grad_norm": 1.0239830343468235,
"learning_rate": 9.465286353846905e-06,
"loss": 0.2108,
"step": 190
},
{
"epoch": 0.49504950495049505,
"grad_norm": 0.6639864769738921,
"learning_rate": 9.36393091047441e-06,
"loss": 0.2096,
"step": 200
},
{
"epoch": 0.5198019801980198,
"grad_norm": 0.7442256820961616,
"learning_rate": 9.254427651223434e-06,
"loss": 0.2096,
"step": 210
},
{
"epoch": 0.5445544554455446,
"grad_norm": 0.6200758715629883,
"learning_rate": 9.136981027633834e-06,
"loss": 0.2088,
"step": 220
},
{
"epoch": 0.5693069306930693,
"grad_norm": 0.6806953465136265,
"learning_rate": 9.011810322156269e-06,
"loss": 0.2188,
"step": 230
},
{
"epoch": 0.594059405940594,
"grad_norm": 0.5965222339817741,
"learning_rate": 8.879149238733932e-06,
"loss": 0.2122,
"step": 240
},
{
"epoch": 0.6188118811881188,
"grad_norm": 0.5937778347833192,
"learning_rate": 8.739245466458187e-06,
"loss": 0.198,
"step": 250
},
{
"epoch": 0.6435643564356436,
"grad_norm": 0.6330113114538947,
"learning_rate": 8.592360217112759e-06,
"loss": 0.2014,
"step": 260
},
{
"epoch": 0.6683168316831684,
"grad_norm": 0.6647979286799263,
"learning_rate": 8.438767737469995e-06,
"loss": 0.1965,
"step": 270
},
{
"epoch": 0.693069306930693,
"grad_norm": 0.5972574353654555,
"learning_rate": 8.278754797249702e-06,
"loss": 0.1936,
"step": 280
},
{
"epoch": 0.7178217821782178,
"grad_norm": 0.6961534332272147,
"learning_rate": 8.11262015369663e-06,
"loss": 0.1983,
"step": 290
},
{
"epoch": 0.7425742574257426,
"grad_norm": 0.5890977053667193,
"learning_rate": 7.940673993776258e-06,
"loss": 0.2124,
"step": 300
},
{
"epoch": 0.7673267326732673,
"grad_norm": 0.5782056754231403,
"learning_rate": 7.763237355030384e-06,
"loss": 0.1912,
"step": 310
},
{
"epoch": 0.7920792079207921,
"grad_norm": 0.47426904980129864,
"learning_rate": 7.580641526173758e-06,
"loss": 0.1998,
"step": 320
},
{
"epoch": 0.8168316831683168,
"grad_norm": 0.6336424200096511,
"learning_rate": 7.39322742855097e-06,
"loss": 0.2092,
"step": 330
},
{
"epoch": 0.8415841584158416,
"grad_norm": 0.5851430863896525,
"learning_rate": 7.201344979608423e-06,
"loss": 0.2048,
"step": 340
},
{
"epoch": 0.8663366336633663,
"grad_norm": 0.5985972703165557,
"learning_rate": 7.0053524395698345e-06,
"loss": 0.1997,
"step": 350
},
{
"epoch": 0.8910891089108911,
"grad_norm": 0.5359885389422082,
"learning_rate": 6.805615742535117e-06,
"loss": 0.1924,
"step": 360
},
{
"epoch": 0.9158415841584159,
"grad_norm": 0.6513367769509769,
"learning_rate": 6.602507813251478e-06,
"loss": 0.1962,
"step": 370
},
{
"epoch": 0.9405940594059405,
"grad_norm": 2.5495071867624897,
"learning_rate": 6.396407870832419e-06,
"loss": 0.1916,
"step": 380
},
{
"epoch": 0.9653465346534653,
"grad_norm": 0.533464536541958,
"learning_rate": 6.187700720724648e-06,
"loss": 0.1867,
"step": 390
},
{
"epoch": 0.9900990099009901,
"grad_norm": 0.6152123883544872,
"learning_rate": 5.976776036244833e-06,
"loss": 0.1932,
"step": 400
},
{
"epoch": 1.0148514851485149,
"grad_norm": 0.5754905609165137,
"learning_rate": 5.764027631027659e-06,
"loss": 0.1528,
"step": 410
},
{
"epoch": 1.0396039603960396,
"grad_norm": 0.6609360008628948,
"learning_rate": 5.549852723743564e-06,
"loss": 0.1244,
"step": 420
},
{
"epoch": 1.0643564356435644,
"grad_norm": 0.6141937179311839,
"learning_rate": 5.334651196459003e-06,
"loss": 0.1261,
"step": 430
},
{
"epoch": 1.0891089108910892,
"grad_norm": 0.6889260069385835,
"learning_rate": 5.118824848023926e-06,
"loss": 0.1173,
"step": 440
},
{
"epoch": 1.113861386138614,
"grad_norm": 0.8267827558845927,
"learning_rate": 4.902776643880461e-06,
"loss": 0.1248,
"step": 450
},
{
"epoch": 1.1386138613861387,
"grad_norm": 0.5753472949830176,
"learning_rate": 4.686909963693498e-06,
"loss": 0.1171,
"step": 460
},
{
"epoch": 1.1633663366336633,
"grad_norm": 0.6505293177551238,
"learning_rate": 4.47162784820784e-06,
"loss": 0.1187,
"step": 470
},
{
"epoch": 1.188118811881188,
"grad_norm": 0.6728733056608579,
"learning_rate": 4.257332246738201e-06,
"loss": 0.1188,
"step": 480
},
{
"epoch": 1.2128712871287128,
"grad_norm": 0.6334415322292648,
"learning_rate": 4.04442326669695e-06,
"loss": 0.1211,
"step": 490
},
{
"epoch": 1.2376237623762376,
"grad_norm": 0.6692215992603302,
"learning_rate": 3.833298426560851e-06,
"loss": 0.118,
"step": 500
},
{
"epoch": 1.2623762376237624,
"grad_norm": 0.6937673438761942,
"learning_rate": 3.624351913671571e-06,
"loss": 0.1181,
"step": 510
},
{
"epoch": 1.2871287128712872,
"grad_norm": 0.7260772952384915,
"learning_rate": 3.4179738482556648e-06,
"loss": 0.1222,
"step": 520
},
{
"epoch": 1.311881188118812,
"grad_norm": 0.7018430210814361,
"learning_rate": 3.214549555038218e-06,
"loss": 0.119,
"step": 530
},
{
"epoch": 1.3366336633663367,
"grad_norm": 0.6398785057605294,
"learning_rate": 3.0144588438100693e-06,
"loss": 0.1169,
"step": 540
},
{
"epoch": 1.3613861386138613,
"grad_norm": 0.6102680360344789,
"learning_rate": 2.8180753002918735e-06,
"loss": 0.1172,
"step": 550
},
{
"epoch": 1.386138613861386,
"grad_norm": 0.6214065811958993,
"learning_rate": 2.6257655886190147e-06,
"loss": 0.1097,
"step": 560
},
{
"epoch": 1.4108910891089108,
"grad_norm": 0.6069249122730125,
"learning_rate": 2.4378887667496696e-06,
"loss": 0.1167,
"step": 570
},
{
"epoch": 1.4356435643564356,
"grad_norm": 0.6906365846153183,
"learning_rate": 2.2547956160742473e-06,
"loss": 0.1168,
"step": 580
},
{
"epoch": 1.4603960396039604,
"grad_norm": 0.6780444546407847,
"learning_rate": 2.0768279864778475e-06,
"loss": 0.1263,
"step": 590
},
{
"epoch": 1.4851485148514851,
"grad_norm": 0.6276144596350219,
"learning_rate": 1.9043181580785597e-06,
"loss": 0.1125,
"step": 600
},
{
"epoch": 1.50990099009901,
"grad_norm": 0.8105107613665639,
"learning_rate": 1.73758822083332e-06,
"loss": 0.1183,
"step": 610
},
{
"epoch": 1.5346534653465347,
"grad_norm": 0.593668139214893,
"learning_rate": 1.5769494731696206e-06,
"loss": 0.1108,
"step": 620
},
{
"epoch": 1.5594059405940595,
"grad_norm": 0.7019089646578581,
"learning_rate": 1.4227018407658822e-06,
"loss": 0.1212,
"step": 630
},
{
"epoch": 1.5841584158415842,
"grad_norm": 0.6425882205315401,
"learning_rate": 1.275133316565691e-06,
"loss": 0.1184,
"step": 640
},
{
"epoch": 1.608910891089109,
"grad_norm": 0.5670206419588937,
"learning_rate": 1.1345194230714235e-06,
"loss": 0.1069,
"step": 650
},
{
"epoch": 1.6336633663366338,
"grad_norm": 0.761904942958332,
"learning_rate": 1.001122697921197e-06,
"loss": 0.1175,
"step": 660
},
{
"epoch": 1.6584158415841586,
"grad_norm": 0.6045327206848765,
"learning_rate": 8.751922037096328e-07,
"loss": 0.1129,
"step": 670
},
{
"epoch": 1.6831683168316833,
"grad_norm": 0.6977601855593518,
"learning_rate": 7.569630629676294e-07,
"loss": 0.1151,
"step": 680
},
{
"epoch": 1.7079207920792079,
"grad_norm": 0.5892491224169939,
"learning_rate": 6.466560191693566e-07,
"loss": 0.1114,
"step": 690
},
{
"epoch": 1.7326732673267327,
"grad_norm": 0.6349779012742225,
"learning_rate": 5.444770245861553e-07,
"loss": 0.1062,
"step": 700
},
{
"epoch": 1.7574257425742574,
"grad_norm": 0.5280424120170226,
"learning_rate": 4.506168557567886e-07,
"loss": 0.1101,
"step": 710
},
{
"epoch": 1.7821782178217822,
"grad_norm": 0.6082009840024847,
"learning_rate": 3.6525075729205274e-07,
"loss": 0.1076,
"step": 720
},
{
"epoch": 1.806930693069307,
"grad_norm": 0.7416157250316486,
"learning_rate": 2.8853811467875413e-07,
"loss": 0.1156,
"step": 730
},
{
"epoch": 1.8316831683168315,
"grad_norm": 0.6796922655628035,
"learning_rate": 2.2062215669397201e-07,
"loss": 0.1199,
"step": 740
},
{
"epoch": 1.8564356435643563,
"grad_norm": 0.519247955113028,
"learning_rate": 1.616296879852175e-07,
"loss": 0.1063,
"step": 750
},
{
"epoch": 1.881188118811881,
"grad_norm": 0.6685776260821267,
"learning_rate": 1.1167085231579111e-07,
"loss": 0.1175,
"step": 760
},
{
"epoch": 1.9059405940594059,
"grad_norm": 0.8282713027884988,
"learning_rate": 7.083892691736428e-08,
"loss": 0.1196,
"step": 770
},
{
"epoch": 1.9306930693069306,
"grad_norm": 0.8408881587671665,
"learning_rate": 3.9210148333763135e-08,
"loss": 0.1081,
"step": 780
},
{
"epoch": 1.9554455445544554,
"grad_norm": 0.6283575387075083,
"learning_rate": 1.684357008110593e-08,
"loss": 0.1135,
"step": 790
},
{
"epoch": 1.9801980198019802,
"grad_norm": 0.6309254787259336,
"learning_rate": 3.780952390058379e-09,
"loss": 0.105,
"step": 800
}
],
"logging_steps": 10,
"max_steps": 808,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.38221627375616e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}