swift/MS-LongWriter-Qwen2-7B-Instruct
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: swift/MS-LongWriter-Qwen2-7B-Instruct
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-11 12:19:13 +08:00
commit 89c3260676
22 changed files with 456268 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

930
trainer_state.json Normal file
View File

@@ -0,0 +1,930 @@
{
"best_metric": 0.59626758,
"best_model_checkpoint": "/mnt/workspace/xingjun.wxj/swift_work/swift-by/output/qwen2-7b-instruct/v7-20240904-182455/checkpoint-432",
"epoch": 1.9965337954939342,
"eval_steps": 200,
"global_step": 432,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"acc": 0.81453747,
"epoch": 0.00462160600808781,
"grad_norm": 2.1875,
"learning_rate": 2.2727272727272729e-07,
"loss": 0.76972306,
"memory(GiB)": 94.61,
"step": 1,
"train_speed(iter/s)": 0.038737
},
{
"acc": 0.8124181,
"epoch": 0.023108030040439053,
"grad_norm": 2.6875,
"learning_rate": 1.1363636363636364e-06,
"loss": 0.64217675,
"memory(GiB)": 100.77,
"step": 5,
"train_speed(iter/s)": 0.049551
},
{
"acc": 0.80393114,
"epoch": 0.046216060080878106,
"grad_norm": 2.40625,
"learning_rate": 2.2727272727272728e-06,
"loss": 0.72643595,
"memory(GiB)": 141.03,
"step": 10,
"train_speed(iter/s)": 0.049763
},
{
"acc": 0.79872313,
"epoch": 0.06932409012131716,
"grad_norm": 2.09375,
"learning_rate": 3.409090909090909e-06,
"loss": 0.73066745,
"memory(GiB)": 141.03,
"step": 15,
"train_speed(iter/s)": 0.049428
},
{
"acc": 0.8086524,
"epoch": 0.09243212016175621,
"grad_norm": 2.078125,
"learning_rate": 4.5454545454545455e-06,
"loss": 0.76757164,
"memory(GiB)": 184.34,
"step": 20,
"train_speed(iter/s)": 0.047908
},
{
"acc": 0.80337076,
"epoch": 0.11554015020219527,
"grad_norm": 1.625,
"learning_rate": 5.681818181818183e-06,
"loss": 0.81441174,
"memory(GiB)": 150.97,
"step": 25,
"train_speed(iter/s)": 0.046696
},
{
"acc": 0.8120677,
"epoch": 0.1386481802426343,
"grad_norm": 1.453125,
"learning_rate": 6.818181818181818e-06,
"loss": 0.7133667,
"memory(GiB)": 150.97,
"step": 30,
"train_speed(iter/s)": 0.046543
},
{
"acc": 0.82139053,
"epoch": 0.16175621028307338,
"grad_norm": 1.4765625,
"learning_rate": 7.954545454545455e-06,
"loss": 0.59637189,
"memory(GiB)": 150.98,
"step": 35,
"train_speed(iter/s)": 0.047037
},
{
"acc": 0.81568699,
"epoch": 0.18486424032351242,
"grad_norm": 1.1796875,
"learning_rate": 9.090909090909091e-06,
"loss": 0.70513887,
"memory(GiB)": 150.98,
"step": 40,
"train_speed(iter/s)": 0.046892
},
{
"acc": 0.820224,
"epoch": 0.20797227036395147,
"grad_norm": 1.171875,
"learning_rate": 9.999836101895782e-06,
"loss": 0.69312973,
"memory(GiB)": 150.98,
"step": 45,
"train_speed(iter/s)": 0.046848
},
{
"acc": 0.81471777,
"epoch": 0.23108030040439054,
"grad_norm": 1.234375,
"learning_rate": 9.994100796397954e-06,
"loss": 0.71612887,
"memory(GiB)": 150.98,
"step": 50,
"train_speed(iter/s)": 0.04669
},
{
"acc": 0.81478605,
"epoch": 0.2541883304448296,
"grad_norm": 1.2578125,
"learning_rate": 9.98018132755861e-06,
"loss": 0.64565916,
"memory(GiB)": 150.98,
"step": 55,
"train_speed(iter/s)": 0.046851
},
{
"acc": 0.81756649,
"epoch": 0.2772963604852686,
"grad_norm": 1.28125,
"learning_rate": 9.958100506132127e-06,
"loss": 0.66640944,
"memory(GiB)": 150.98,
"step": 60,
"train_speed(iter/s)": 0.047121
},
{
"acc": 0.81747818,
"epoch": 0.30040439052570767,
"grad_norm": 1.4375,
"learning_rate": 9.927894517421252e-06,
"loss": 0.63872623,
"memory(GiB)": 150.98,
"step": 65,
"train_speed(iter/s)": 0.047364
},
{
"acc": 0.81161108,
"epoch": 0.32351242056614676,
"grad_norm": 1.28125,
"learning_rate": 9.889612861977855e-06,
"loss": 0.7101274,
"memory(GiB)": 150.98,
"step": 70,
"train_speed(iter/s)": 0.047269
},
{
"acc": 0.81774912,
"epoch": 0.3466204506065858,
"grad_norm": 1.3828125,
"learning_rate": 9.843318274483089e-06,
"loss": 0.64770298,
"memory(GiB)": 150.98,
"step": 75,
"train_speed(iter/s)": 0.047298
},
{
"acc": 0.81703777,
"epoch": 0.36972848064702485,
"grad_norm": 1.1328125,
"learning_rate": 9.789086620939936e-06,
"loss": 0.64908137,
"memory(GiB)": 150.98,
"step": 80,
"train_speed(iter/s)": 0.047241
},
{
"acc": 0.8161418,
"epoch": 0.3928365106874639,
"grad_norm": 1.3671875,
"learning_rate": 9.727006774346551e-06,
"loss": 0.66009436,
"memory(GiB)": 150.98,
"step": 85,
"train_speed(iter/s)": 0.047297
},
{
"acc": 0.82398653,
"epoch": 0.41594454072790293,
"grad_norm": 1.3203125,
"learning_rate": 9.657180469054213e-06,
"loss": 0.64000554,
"memory(GiB)": 150.98,
"step": 90,
"train_speed(iter/s)": 0.047467
},
{
"acc": 0.82110977,
"epoch": 0.439052570768342,
"grad_norm": 1.28125,
"learning_rate": 9.579722134048506e-06,
"loss": 0.67605906,
"memory(GiB)": 150.98,
"step": 95,
"train_speed(iter/s)": 0.047599
},
{
"acc": 0.82549,
"epoch": 0.4621606008087811,
"grad_norm": 1.4140625,
"learning_rate": 9.494758705426978e-06,
"loss": 0.58404155,
"memory(GiB)": 150.98,
"step": 100,
"train_speed(iter/s)": 0.047822
},
{
"acc": 0.82027588,
"epoch": 0.4852686308492201,
"grad_norm": 1.28125,
"learning_rate": 9.402429418380555e-06,
"loss": 0.69437795,
"memory(GiB)": 150.98,
"step": 105,
"train_speed(iter/s)": 0.047539
},
{
"acc": 0.82060165,
"epoch": 0.5083766608896592,
"grad_norm": 1.4296875,
"learning_rate": 9.302885579019626e-06,
"loss": 0.60217834,
"memory(GiB)": 150.98,
"step": 110,
"train_speed(iter/s)": 0.047767
},
{
"acc": 0.81884346,
"epoch": 0.5314846909300982,
"grad_norm": 1.40625,
"learning_rate": 9.196290316418712e-06,
"loss": 0.62260752,
"memory(GiB)": 150.98,
"step": 115,
"train_speed(iter/s)": 0.047925
},
{
"acc": 0.81530762,
"epoch": 0.5545927209705372,
"grad_norm": 1.1796875,
"learning_rate": 9.082818315286054e-06,
"loss": 0.67854958,
"memory(GiB)": 150.98,
"step": 120,
"train_speed(iter/s)": 0.048032
},
{
"acc": 0.8216691,
"epoch": 0.5777007510109763,
"grad_norm": 1.28125,
"learning_rate": 8.962655529696236e-06,
"loss": 0.61422577,
"memory(GiB)": 150.98,
"step": 125,
"train_speed(iter/s)": 0.048043
},
{
"acc": 0.82413073,
"epoch": 0.6008087810514153,
"grad_norm": 1.1796875,
"learning_rate": 8.83599887835493e-06,
"loss": 0.59601698,
"memory(GiB)": 150.98,
"step": 130,
"train_speed(iter/s)": 0.048165
},
{
"acc": 0.82421112,
"epoch": 0.6239168110918544,
"grad_norm": 1.3515625,
"learning_rate": 8.7030559218952e-06,
"loss": 0.62850995,
"memory(GiB)": 150.98,
"step": 135,
"train_speed(iter/s)": 0.048186
},
{
"acc": 0.81220446,
"epoch": 0.6470248411322935,
"grad_norm": 1.40625,
"learning_rate": 8.564044522734147e-06,
"loss": 0.65466747,
"memory(GiB)": 150.98,
"step": 140,
"train_speed(iter/s)": 0.048296
},
{
"acc": 0.8225975,
"epoch": 0.6701328711727326,
"grad_norm": 1.3515625,
"learning_rate": 8.419192488047369e-06,
"loss": 0.68851438,
"memory(GiB)": 150.98,
"step": 145,
"train_speed(iter/s)": 0.048275
},
{
"acc": 0.82113705,
"epoch": 0.6932409012131716,
"grad_norm": 1.3203125,
"learning_rate": 8.268737196446264e-06,
"loss": 0.66175132,
"memory(GiB)": 150.98,
"step": 150,
"train_speed(iter/s)": 0.048236
},
{
"acc": 0.82547045,
"epoch": 0.7163489312536107,
"grad_norm": 1.1875,
"learning_rate": 8.112925208969996e-06,
"loss": 0.58082228,
"memory(GiB)": 150.98,
"step": 155,
"train_speed(iter/s)": 0.048304
},
{
"acc": 0.81520081,
"epoch": 0.7394569612940497,
"grad_norm": 1.15625,
"learning_rate": 7.952011865029614e-06,
"loss": 0.7566462,
"memory(GiB)": 150.98,
"step": 160,
"train_speed(iter/s)": 0.04806
},
{
"acc": 0.82244606,
"epoch": 0.7625649913344887,
"grad_norm": 1.265625,
"learning_rate": 7.786260863966468e-06,
"loss": 0.64169059,
"memory(GiB)": 150.98,
"step": 165,
"train_speed(iter/s)": 0.048128
},
{
"acc": 0.82460566,
"epoch": 0.7856730213749278,
"grad_norm": 1.203125,
"learning_rate": 7.61594383291065e-06,
"loss": 0.64750757,
"memory(GiB)": 150.98,
"step": 170,
"train_speed(iter/s)": 0.048108
},
{
"acc": 0.82297211,
"epoch": 0.8087810514153668,
"grad_norm": 1.1484375,
"learning_rate": 7.441339881647654e-06,
"loss": 0.62678857,
"memory(GiB)": 150.98,
"step": 175,
"train_speed(iter/s)": 0.048123
},
{
"acc": 0.82202625,
"epoch": 0.8318890814558059,
"grad_norm": 1.2109375,
"learning_rate": 7.262735145222696e-06,
"loss": 0.63359532,
"memory(GiB)": 150.98,
"step": 180,
"train_speed(iter/s)": 0.04815
},
{
"acc": 0.81521053,
"epoch": 0.8549971114962449,
"grad_norm": 1.3984375,
"learning_rate": 7.080422315032298e-06,
"loss": 0.66546955,
"memory(GiB)": 150.98,
"step": 185,
"train_speed(iter/s)": 0.048203
},
{
"acc": 0.83204832,
"epoch": 0.878105141536684,
"grad_norm": 1.3203125,
"learning_rate": 6.894700159171535e-06,
"loss": 0.60190206,
"memory(GiB)": 150.98,
"step": 190,
"train_speed(iter/s)": 0.04822
},
{
"acc": 0.81509237,
"epoch": 0.901213171577123,
"grad_norm": 1.15625,
"learning_rate": 6.705873032822981e-06,
"loss": 0.71670413,
"memory(GiB)": 150.98,
"step": 195,
"train_speed(iter/s)": 0.048127
},
{
"acc": 0.82601328,
"epoch": 0.9243212016175621,
"grad_norm": 1.3984375,
"learning_rate": 6.514250379489754e-06,
"loss": 0.55743012,
"memory(GiB)": 150.98,
"step": 200,
"train_speed(iter/s)": 0.048221
},
{
"epoch": 0.9243212016175621,
"eval_acc": 0.7485949746717407,
"eval_loss": 0.6565380096435547,
"eval_runtime": 11.0891,
"eval_samples_per_second": 12.445,
"eval_steps_per_second": 12.445,
"step": 200
},
{
"acc": 0.82308855,
"epoch": 0.9474292316580012,
"grad_norm": 1.4375,
"learning_rate": 6.320146223889966e-06,
"loss": 0.63252063,
"memory(GiB)": 150.98,
"step": 205,
"train_speed(iter/s)": 0.046847
},
{
"acc": 0.82361374,
"epoch": 0.9705372616984402,
"grad_norm": 1.421875,
"learning_rate": 6.123878657343648e-06,
"loss": 0.59042826,
"memory(GiB)": 150.98,
"step": 210,
"train_speed(iter/s)": 0.046942
},
{
"acc": 0.82532425,
"epoch": 0.9936452917388793,
"grad_norm": 0.98828125,
"learning_rate": 5.925769316495461e-06,
"loss": 0.60934105,
"memory(GiB)": 150.98,
"step": 215,
"train_speed(iter/s)": 0.046992
},
{
"acc": 0.83245983,
"epoch": 1.0167533217793183,
"grad_norm": 1.1953125,
"learning_rate": 5.726142856227453e-06,
"loss": 0.65204225,
"memory(GiB)": 150.98,
"step": 220,
"train_speed(iter/s)": 0.046872
},
{
"acc": 0.83649168,
"epoch": 1.0398613518197575,
"grad_norm": 1.2109375,
"learning_rate": 5.52532641762562e-06,
"loss": 0.66422782,
"memory(GiB)": 150.98,
"step": 225,
"train_speed(iter/s)": 0.046806
},
{
"acc": 0.83163223,
"epoch": 1.0629693818601964,
"grad_norm": 1.140625,
"learning_rate": 5.323649091872179e-06,
"loss": 0.60072503,
"memory(GiB)": 150.98,
"step": 230,
"train_speed(iter/s)": 0.046824
},
{
"acc": 0.83812866,
"epoch": 1.0860774119006356,
"grad_norm": 1.25,
"learning_rate": 5.121441380942066e-06,
"loss": 0.58486128,
"memory(GiB)": 150.98,
"step": 235,
"train_speed(iter/s)": 0.046865
},
{
"acc": 0.82630444,
"epoch": 1.1091854419410745,
"grad_norm": 1.1328125,
"learning_rate": 4.919034655987493e-06,
"loss": 0.64795966,
"memory(GiB)": 150.98,
"step": 240,
"train_speed(iter/s)": 0.046852
},
{
"acc": 0.83698711,
"epoch": 1.1322934719815136,
"grad_norm": 1.4375,
"learning_rate": 4.716760614298118e-06,
"loss": 0.5604661,
"memory(GiB)": 150.98,
"step": 245,
"train_speed(iter/s)": 0.046935
},
{
"acc": 0.82200708,
"epoch": 1.1554015020219526,
"grad_norm": 1.2265625,
"learning_rate": 4.51495073572676e-06,
"loss": 0.65722399,
"memory(GiB)": 150.98,
"step": 250,
"train_speed(iter/s)": 0.046964
},
{
"acc": 0.83468876,
"epoch": 1.1785095320623917,
"grad_norm": 1.125,
"learning_rate": 4.313935739471425e-06,
"loss": 0.58888826,
"memory(GiB)": 150.98,
"step": 255,
"train_speed(iter/s)": 0.046962
},
{
"acc": 0.82915096,
"epoch": 1.2016175621028307,
"grad_norm": 1.4296875,
"learning_rate": 4.1140450421038865e-06,
"loss": 0.64913301,
"memory(GiB)": 150.98,
"step": 260,
"train_speed(iter/s)": 0.04701
},
{
"acc": 0.8357687,
"epoch": 1.2247255921432698,
"grad_norm": 1.2578125,
"learning_rate": 3.9156062177329624e-06,
"loss": 0.58918366,
"memory(GiB)": 150.98,
"step": 265,
"train_speed(iter/s)": 0.04707
},
{
"acc": 0.83036833,
"epoch": 1.2478336221837087,
"grad_norm": 1.203125,
"learning_rate": 3.7189444611871383e-06,
"loss": 0.60051751,
"memory(GiB)": 150.98,
"step": 270,
"train_speed(iter/s)": 0.047112
},
{
"acc": 0.8339922,
"epoch": 1.270941652224148,
"grad_norm": 1.3671875,
"learning_rate": 3.524382055096309e-06,
"loss": 0.55478253,
"memory(GiB)": 150.98,
"step": 275,
"train_speed(iter/s)": 0.047186
},
{
"acc": 0.83721056,
"epoch": 1.2940496822645868,
"grad_norm": 1.1328125,
"learning_rate": 3.3322378417458985e-06,
"loss": 0.59262004,
"memory(GiB)": 150.98,
"step": 280,
"train_speed(iter/s)": 0.047233
},
{
"acc": 0.82519064,
"epoch": 1.317157712305026,
"grad_norm": 1.171875,
"learning_rate": 3.1428267005689182e-06,
"loss": 0.68231759,
"memory(GiB)": 150.98,
"step": 285,
"train_speed(iter/s)": 0.047214
},
{
"acc": 0.83670778,
"epoch": 1.3402657423454651,
"grad_norm": 1.203125,
"learning_rate": 2.9564590321322206e-06,
"loss": 0.54285073,
"memory(GiB)": 150.98,
"step": 290,
"train_speed(iter/s)": 0.047272
},
{
"acc": 0.8335681,
"epoch": 1.363373772385904,
"grad_norm": 1.1953125,
"learning_rate": 2.7734402494625395e-06,
"loss": 0.60112472,
"memory(GiB)": 150.98,
"step": 295,
"train_speed(iter/s)": 0.04725
},
{
"acc": 0.83400679,
"epoch": 1.3864818024263432,
"grad_norm": 1.1328125,
"learning_rate": 2.594070277545975e-06,
"loss": 0.61252079,
"memory(GiB)": 150.98,
"step": 300,
"train_speed(iter/s)": 0.047279
},
{
"acc": 0.83710117,
"epoch": 1.4095898324667822,
"grad_norm": 1.2109375,
"learning_rate": 2.4186430618210704e-06,
"loss": 0.53748951,
"memory(GiB)": 150.98,
"step": 305,
"train_speed(iter/s)": 0.047266
},
{
"acc": 0.8276473,
"epoch": 1.4326978625072213,
"grad_norm": 1.3125,
"learning_rate": 2.2474460864709825e-06,
"loss": 0.6252665,
"memory(GiB)": 150.98,
"step": 310,
"train_speed(iter/s)": 0.04732
},
{
"acc": 0.83237705,
"epoch": 1.4558058925476602,
"grad_norm": 1.1328125,
"learning_rate": 2.0807599033041238e-06,
"loss": 0.63391018,
"memory(GiB)": 150.98,
"step": 315,
"train_speed(iter/s)": 0.04733
},
{
"acc": 0.83387976,
"epoch": 1.4789139225880994,
"grad_norm": 1.046875,
"learning_rate": 1.9188576719953635e-06,
"loss": 0.63072858,
"memory(GiB)": 150.98,
"step": 320,
"train_speed(iter/s)": 0.047302
},
{
"acc": 0.83439655,
"epoch": 1.5020219526285383,
"grad_norm": 1.2265625,
"learning_rate": 1.7620047124412e-06,
"loss": 0.615345,
"memory(GiB)": 150.98,
"step": 325,
"train_speed(iter/s)": 0.04731
},
{
"acc": 0.83732977,
"epoch": 1.5251299826689775,
"grad_norm": 1.3671875,
"learning_rate": 1.6104580699624839e-06,
"loss": 0.57706161,
"memory(GiB)": 150.98,
"step": 330,
"train_speed(iter/s)": 0.04735
},
{
"acc": 0.82502375,
"epoch": 1.5482380127094166,
"grad_norm": 1.2578125,
"learning_rate": 1.4644660940672628e-06,
"loss": 0.68620262,
"memory(GiB)": 150.98,
"step": 335,
"train_speed(iter/s)": 0.047298
},
{
"acc": 0.83036242,
"epoch": 1.5713460427498556,
"grad_norm": 1.1015625,
"learning_rate": 1.3242680314639995e-06,
"loss": 0.65163517,
"memory(GiB)": 150.98,
"step": 340,
"train_speed(iter/s)": 0.047314
},
{
"acc": 0.82439814,
"epoch": 1.5944540727902945,
"grad_norm": 1.140625,
"learning_rate": 1.1900936339921692e-06,
"loss": 0.64395781,
"memory(GiB)": 150.98,
"step": 345,
"train_speed(iter/s)": 0.04733
},
{
"acc": 0.83437843,
"epoch": 1.6175621028307337,
"grad_norm": 1.28125,
"learning_rate": 1.062162782112729e-06,
"loss": 0.58336234,
"memory(GiB)": 150.98,
"step": 350,
"train_speed(iter/s)": 0.047379
},
{
"acc": 0.83543835,
"epoch": 1.6406701328711728,
"grad_norm": 1.1640625,
"learning_rate": 9.406851245754478e-07,
"loss": 0.6100008,
"memory(GiB)": 150.98,
"step": 355,
"train_speed(iter/s)": 0.047394
},
{
"acc": 0.82420549,
"epoch": 1.6637781629116117,
"grad_norm": 1.3515625,
"learning_rate": 8.258597348536452e-07,
"loss": 0.67773538,
"memory(GiB)": 150.98,
"step": 360,
"train_speed(iter/s)": 0.047419
},
{
"acc": 0.83056965,
"epoch": 1.686886192952051,
"grad_norm": 1.1484375,
"learning_rate": 7.178747849093093e-07,
"loss": 0.63277826,
"memory(GiB)": 150.98,
"step": 365,
"train_speed(iter/s)": 0.047439
},
{
"acc": 0.83617363,
"epoch": 1.7099942229924898,
"grad_norm": 1.28125,
"learning_rate": 6.16907236823262e-07,
"loss": 0.58432493,
"memory(GiB)": 150.98,
"step": 370,
"train_speed(iter/s)": 0.04747
},
{
"acc": 0.82903013,
"epoch": 1.733102253032929,
"grad_norm": 1.1640625,
"learning_rate": 5.231225527956924e-07,
"loss": 0.6580874,
"memory(GiB)": 150.98,
"step": 375,
"train_speed(iter/s)": 0.047437
},
{
"acc": 0.83327017,
"epoch": 1.7562102830733681,
"grad_norm": 1.1328125,
"learning_rate": 4.3667442399229985e-07,
"loss": 0.58462934,
"memory(GiB)": 150.98,
"step": 380,
"train_speed(iter/s)": 0.047468
},
{
"acc": 0.83651209,
"epoch": 1.779318313113807,
"grad_norm": 1.125,
"learning_rate": 3.577045186804118e-07,
"loss": 0.58018489,
"memory(GiB)": 150.98,
"step": 385,
"train_speed(iter/s)": 0.047511
},
{
"acc": 0.82789068,
"epoch": 1.802426343154246,
"grad_norm": 1.0703125,
"learning_rate": 2.8634225006782867e-07,
"loss": 0.67336974,
"memory(GiB)": 150.98,
"step": 390,
"train_speed(iter/s)": 0.047483
},
{
"acc": 0.84120426,
"epoch": 1.8255343731946851,
"grad_norm": 1.2109375,
"learning_rate": 2.2270456422483655e-07,
"loss": 0.58580966,
"memory(GiB)": 150.98,
"step": 395,
"train_speed(iter/s)": 0.047495
},
{
"acc": 0.82003698,
"epoch": 1.8486424032351243,
"grad_norm": 1.125,
"learning_rate": 1.6689574843694433e-07,
"loss": 0.77600121,
"memory(GiB)": 150.98,
"step": 400,
"train_speed(iter/s)": 0.04747
},
{
"epoch": 1.8486424032351243,
"eval_acc": 0.7489352050202878,
"eval_loss": 0.6084072589874268,
"eval_runtime": 11.0063,
"eval_samples_per_second": 12.538,
"eval_steps_per_second": 12.538,
"step": 400
},
{
"acc": 0.83795338,
"epoch": 1.8717504332755632,
"grad_norm": 1.3671875,
"learning_rate": 1.1900726030241005e-07,
"loss": 0.57120104,
"memory(GiB)": 150.98,
"step": 405,
"train_speed(iter/s)": 0.046796
},
{
"acc": 0.82940121,
"epoch": 1.8948584633160022,
"grad_norm": 1.2109375,
"learning_rate": 7.911757785462882e-08,
"loss": 0.63004894,
"memory(GiB)": 150.98,
"step": 410,
"train_speed(iter/s)": 0.046799
},
{
"acc": 0.82601376,
"epoch": 1.9179664933564413,
"grad_norm": 1.3125,
"learning_rate": 4.729207095498345e-08,
"loss": 0.5973454,
"memory(GiB)": 150.98,
"step": 415,
"train_speed(iter/s)": 0.046849
},
{
"acc": 0.83928432,
"epoch": 1.9410745233968805,
"grad_norm": 1.2578125,
"learning_rate": 2.358289416693027e-08,
"loss": 0.57047229,
"memory(GiB)": 150.98,
"step": 420,
"train_speed(iter/s)": 0.046875
},
{
"acc": 0.83347263,
"epoch": 1.9641825534373196,
"grad_norm": 1.1171875,
"learning_rate": 8.02890128686562e-09,
"loss": 0.56035862,
"memory(GiB)": 150.98,
"step": 425,
"train_speed(iter/s)": 0.046877
},
{
"acc": 0.83463163,
"epoch": 1.9872905834777586,
"grad_norm": 1.28125,
"learning_rate": 6.555816718389896e-10,
"loss": 0.57831531,
"memory(GiB)": 150.98,
"step": 430,
"train_speed(iter/s)": 0.046917
},
{
"epoch": 1.9965337954939342,
"eval_acc": 0.7487839915320447,
"eval_loss": 0.596267580986023,
"eval_runtime": 11.0681,
"eval_samples_per_second": 12.468,
"eval_steps_per_second": 12.468,
"step": 432
}
],
"logging_steps": 5,
"max_steps": 432,
"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": true
},
"attributes": {}
}
},
"total_flos": 8.150084761585428e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}