Yaseal/llama3_3b_instruct_vallina_full_sft_30k
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: Yaseal/llama3_3b_instruct_vallina_full_sft_30k
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-04-11 23:15:17 +08:00
commit 2e0befc9df
258 changed files with 53691 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

863
trainer_state.json Normal file
View File

@@ -0,0 +1,863 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 1000,
"global_step": 1164,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01718213058419244,
"grad_norm": 1.9297058582305908,
"learning_rate": 1.5384615384615387e-06,
"loss": 0.8691,
"step": 10
},
{
"epoch": 0.03436426116838488,
"grad_norm": 1.0176167488098145,
"learning_rate": 3.247863247863248e-06,
"loss": 0.7729,
"step": 20
},
{
"epoch": 0.05154639175257732,
"grad_norm": 0.7100021839141846,
"learning_rate": 4.957264957264958e-06,
"loss": 0.6791,
"step": 30
},
{
"epoch": 0.06872852233676977,
"grad_norm": 0.5990766882896423,
"learning_rate": 6.666666666666667e-06,
"loss": 0.6453,
"step": 40
},
{
"epoch": 0.0859106529209622,
"grad_norm": 0.5850998759269714,
"learning_rate": 8.376068376068377e-06,
"loss": 0.612,
"step": 50
},
{
"epoch": 0.10309278350515463,
"grad_norm": 0.5561569929122925,
"learning_rate": 1.0085470085470086e-05,
"loss": 0.6023,
"step": 60
},
{
"epoch": 0.12027491408934708,
"grad_norm": 0.6188008785247803,
"learning_rate": 1.1794871794871796e-05,
"loss": 0.5756,
"step": 70
},
{
"epoch": 0.13745704467353953,
"grad_norm": 0.575279712677002,
"learning_rate": 1.3504273504273506e-05,
"loss": 0.5798,
"step": 80
},
{
"epoch": 0.15463917525773196,
"grad_norm": 0.5770293474197388,
"learning_rate": 1.5213675213675214e-05,
"loss": 0.5444,
"step": 90
},
{
"epoch": 0.1718213058419244,
"grad_norm": 0.5948958992958069,
"learning_rate": 1.6923076923076924e-05,
"loss": 0.5604,
"step": 100
},
{
"epoch": 0.18900343642611683,
"grad_norm": 0.6553467512130737,
"learning_rate": 1.8632478632478636e-05,
"loss": 0.5569,
"step": 110
},
{
"epoch": 0.20618556701030927,
"grad_norm": 0.6900191307067871,
"learning_rate": 1.999981993260809e-05,
"loss": 0.5541,
"step": 120
},
{
"epoch": 0.22336769759450173,
"grad_norm": 0.604987382888794,
"learning_rate": 1.9993518254774517e-05,
"loss": 0.5664,
"step": 130
},
{
"epoch": 0.24054982817869416,
"grad_norm": 0.6313670873641968,
"learning_rate": 1.9978219691148676e-05,
"loss": 0.5433,
"step": 140
},
{
"epoch": 0.25773195876288657,
"grad_norm": 0.5558223128318787,
"learning_rate": 1.9953938014600757e-05,
"loss": 0.5348,
"step": 150
},
{
"epoch": 0.27491408934707906,
"grad_norm": 0.5769678950309753,
"learning_rate": 1.9920695085247012e-05,
"loss": 0.5427,
"step": 160
},
{
"epoch": 0.2920962199312715,
"grad_norm": 0.5733423233032227,
"learning_rate": 1.9878520830769675e-05,
"loss": 0.5441,
"step": 170
},
{
"epoch": 0.30927835051546393,
"grad_norm": 0.6078853607177734,
"learning_rate": 1.9827453219473925e-05,
"loss": 0.5425,
"step": 180
},
{
"epoch": 0.32646048109965636,
"grad_norm": 0.595964789390564,
"learning_rate": 1.9767538226106078e-05,
"loss": 0.5386,
"step": 190
},
{
"epoch": 0.3436426116838488,
"grad_norm": 0.6116155385971069,
"learning_rate": 1.9698829790463792e-05,
"loss": 0.54,
"step": 200
},
{
"epoch": 0.36082474226804123,
"grad_norm": 0.6045894026756287,
"learning_rate": 1.962138976883558e-05,
"loss": 0.5302,
"step": 210
},
{
"epoch": 0.37800687285223367,
"grad_norm": 0.5803776979446411,
"learning_rate": 1.9535287878313315e-05,
"loss": 0.5323,
"step": 220
},
{
"epoch": 0.3951890034364261,
"grad_norm": 0.619290292263031,
"learning_rate": 1.9440601634027892e-05,
"loss": 0.5194,
"step": 230
},
{
"epoch": 0.41237113402061853,
"grad_norm": 0.554390013217926,
"learning_rate": 1.9337416279364486e-05,
"loss": 0.5298,
"step": 240
},
{
"epoch": 0.42955326460481097,
"grad_norm": 0.5785292387008667,
"learning_rate": 1.922582470922034e-05,
"loss": 0.5246,
"step": 250
},
{
"epoch": 0.44673539518900346,
"grad_norm": 0.5407880544662476,
"learning_rate": 1.910592738637407e-05,
"loss": 0.5066,
"step": 260
},
{
"epoch": 0.4639175257731959,
"grad_norm": 0.5649763345718384,
"learning_rate": 1.897783225104181e-05,
"loss": 0.5303,
"step": 270
},
{
"epoch": 0.48109965635738833,
"grad_norm": 0.5799049139022827,
"learning_rate": 1.8841654623701673e-05,
"loss": 0.5225,
"step": 280
},
{
"epoch": 0.49828178694158076,
"grad_norm": 0.581510066986084,
"learning_rate": 1.869751710127392e-05,
"loss": 0.5052,
"step": 290
},
{
"epoch": 0.5154639175257731,
"grad_norm": 0.587986946105957,
"learning_rate": 1.8545549446750392e-05,
"loss": 0.5143,
"step": 300
},
{
"epoch": 0.5326460481099656,
"grad_norm": 0.5544498562812805,
"learning_rate": 1.8385888472372474e-05,
"loss": 0.5129,
"step": 310
},
{
"epoch": 0.5498281786941581,
"grad_norm": 0.5483389496803284,
"learning_rate": 1.8218677916462882e-05,
"loss": 0.5131,
"step": 320
},
{
"epoch": 0.5670103092783505,
"grad_norm": 0.5526708960533142,
"learning_rate": 1.8044068314022057e-05,
"loss": 0.5099,
"step": 330
},
{
"epoch": 0.584192439862543,
"grad_norm": 0.5597785115242004,
"learning_rate": 1.786221686120567e-05,
"loss": 0.5166,
"step": 340
},
{
"epoch": 0.6013745704467354,
"grad_norm": 0.5774782299995422,
"learning_rate": 1.7673287273805342e-05,
"loss": 0.5228,
"step": 350
},
{
"epoch": 0.6185567010309279,
"grad_norm": 0.6076422929763794,
"learning_rate": 1.7477449639859876e-05,
"loss": 0.5099,
"step": 360
},
{
"epoch": 0.6357388316151202,
"grad_norm": 0.5519262552261353,
"learning_rate": 1.7274880266529716e-05,
"loss": 0.5047,
"step": 370
},
{
"epoch": 0.6529209621993127,
"grad_norm": 0.5036320686340332,
"learning_rate": 1.7065761521372562e-05,
"loss": 0.4984,
"step": 380
},
{
"epoch": 0.6701030927835051,
"grad_norm": 0.523205578327179,
"learning_rate": 1.6850281668162908e-05,
"loss": 0.5079,
"step": 390
},
{
"epoch": 0.6872852233676976,
"grad_norm": 0.5750735998153687,
"learning_rate": 1.6628634697403447e-05,
"loss": 0.5086,
"step": 400
},
{
"epoch": 0.7044673539518901,
"grad_norm": 0.541567325592041,
"learning_rate": 1.6401020151680815e-05,
"loss": 0.4957,
"step": 410
},
{
"epoch": 0.7216494845360825,
"grad_norm": 0.5553820133209229,
"learning_rate": 1.6167642946022956e-05,
"loss": 0.5021,
"step": 420
},
{
"epoch": 0.738831615120275,
"grad_norm": 0.5656445622444153,
"learning_rate": 1.592871318341986e-05,
"loss": 0.5146,
"step": 430
},
{
"epoch": 0.7560137457044673,
"grad_norm": 0.5366631150245667,
"learning_rate": 1.5684445965673693e-05,
"loss": 0.5087,
"step": 440
},
{
"epoch": 0.7731958762886598,
"grad_norm": 0.5228840708732605,
"learning_rate": 1.5435061199748625e-05,
"loss": 0.5065,
"step": 450
},
{
"epoch": 0.7903780068728522,
"grad_norm": 0.5593631267547607,
"learning_rate": 1.5180783399794749e-05,
"loss": 0.5092,
"step": 460
},
{
"epoch": 0.8075601374570447,
"grad_norm": 0.5312666893005371,
"learning_rate": 1.4921841485024236e-05,
"loss": 0.4999,
"step": 470
},
{
"epoch": 0.8247422680412371,
"grad_norm": 0.48128607869148254,
"learning_rate": 1.4658468573621752e-05,
"loss": 0.4992,
"step": 480
},
{
"epoch": 0.8419243986254296,
"grad_norm": 0.5016392469406128,
"learning_rate": 1.4390901772874668e-05,
"loss": 0.5041,
"step": 490
},
{
"epoch": 0.8591065292096219,
"grad_norm": 0.5364375114440918,
"learning_rate": 1.4119381965711976e-05,
"loss": 0.4942,
"step": 500
},
{
"epoch": 0.8762886597938144,
"grad_norm": 0.5185838937759399,
"learning_rate": 1.3844153593844098e-05,
"loss": 0.4877,
"step": 510
},
{
"epoch": 0.8934707903780069,
"grad_norm": 0.5283859968185425,
"learning_rate": 1.356546443769885e-05,
"loss": 0.4868,
"step": 520
},
{
"epoch": 0.9106529209621993,
"grad_norm": 0.522241473197937,
"learning_rate": 1.328356539335161e-05,
"loss": 0.5083,
"step": 530
},
{
"epoch": 0.9278350515463918,
"grad_norm": 0.5134701132774353,
"learning_rate": 1.2998710246650594e-05,
"loss": 0.4913,
"step": 540
},
{
"epoch": 0.9450171821305842,
"grad_norm": 0.5106884241104126,
"learning_rate": 1.2711155444740529e-05,
"loss": 0.494,
"step": 550
},
{
"epoch": 0.9621993127147767,
"grad_norm": 0.5208893418312073,
"learning_rate": 1.242115986519044e-05,
"loss": 0.5144,
"step": 560
},
{
"epoch": 0.979381443298969,
"grad_norm": 0.5227161049842834,
"learning_rate": 1.2128984582933397e-05,
"loss": 0.4891,
"step": 570
},
{
"epoch": 0.9965635738831615,
"grad_norm": 0.5008633136749268,
"learning_rate": 1.1834892635228024e-05,
"loss": 0.4825,
"step": 580
},
{
"epoch": 1.013745704467354,
"grad_norm": 0.6478720903396606,
"learning_rate": 1.1539148784853404e-05,
"loss": 0.4331,
"step": 590
},
{
"epoch": 1.0309278350515463,
"grad_norm": 0.6044857501983643,
"learning_rate": 1.1242019281750525e-05,
"loss": 0.4127,
"step": 600
},
{
"epoch": 1.0481099656357389,
"grad_norm": 0.5689923167228699,
"learning_rate": 1.0943771623324884e-05,
"loss": 0.4022,
"step": 610
},
{
"epoch": 1.0652920962199313,
"grad_norm": 0.5413344502449036,
"learning_rate": 1.064467431362603e-05,
"loss": 0.4074,
"step": 620
},
{
"epoch": 1.0824742268041236,
"grad_norm": 0.5501529574394226,
"learning_rate": 1.0344996621620862e-05,
"loss": 0.4173,
"step": 630
},
{
"epoch": 1.0996563573883162,
"grad_norm": 0.5490415692329407,
"learning_rate": 1.004500833877828e-05,
"loss": 0.4071,
"step": 640
},
{
"epoch": 1.1168384879725086,
"grad_norm": 0.5644371509552002,
"learning_rate": 9.744979536183458e-06,
"loss": 0.4128,
"step": 650
},
{
"epoch": 1.134020618556701,
"grad_norm": 0.5400048494338989,
"learning_rate": 9.445180321400373e-06,
"loss": 0.4162,
"step": 660
},
{
"epoch": 1.1512027491408934,
"grad_norm": 0.5328329801559448,
"learning_rate": 9.145880595301495e-06,
"loss": 0.4087,
"step": 670
},
{
"epoch": 1.168384879725086,
"grad_norm": 0.5620996952056885,
"learning_rate": 8.847349809083557e-06,
"loss": 0.4104,
"step": 680
},
{
"epoch": 1.1855670103092784,
"grad_norm": 0.5561193823814392,
"learning_rate": 8.54985672168817e-06,
"loss": 0.4186,
"step": 690
},
{
"epoch": 1.2027491408934707,
"grad_norm": 0.5532508492469788,
"learning_rate": 8.253669157845632e-06,
"loss": 0.4027,
"step": 700
},
{
"epoch": 1.2199312714776633,
"grad_norm": 0.5275538563728333,
"learning_rate": 7.959053766959785e-06,
"loss": 0.3996,
"step": 710
},
{
"epoch": 1.2371134020618557,
"grad_norm": 0.5567190647125244,
"learning_rate": 7.666275783051012e-06,
"loss": 0.4053,
"step": 720
},
{
"epoch": 1.254295532646048,
"grad_norm": 0.505807638168335,
"learning_rate": 7.375598785973429e-06,
"loss": 0.4065,
"step": 730
},
{
"epoch": 1.2714776632302405,
"grad_norm": 0.5188184976577759,
"learning_rate": 7.087284464121305e-06,
"loss": 0.4041,
"step": 740
},
{
"epoch": 1.2886597938144329,
"grad_norm": 0.5313155055046082,
"learning_rate": 6.801592378838282e-06,
"loss": 0.4055,
"step": 750
},
{
"epoch": 1.3058419243986255,
"grad_norm": 0.5596489906311035,
"learning_rate": 6.518779730741555e-06,
"loss": 0.4143,
"step": 760
},
{
"epoch": 1.3230240549828178,
"grad_norm": 0.5449512600898743,
"learning_rate": 6.239101128171317e-06,
"loss": 0.3931,
"step": 770
},
{
"epoch": 1.3402061855670104,
"grad_norm": 0.5664635300636292,
"learning_rate": 5.9628083579739715e-06,
"loss": 0.4126,
"step": 780
},
{
"epoch": 1.3573883161512028,
"grad_norm": 0.5540263056755066,
"learning_rate": 5.690150158825462e-06,
"loss": 0.4108,
"step": 790
},
{
"epoch": 1.3745704467353952,
"grad_norm": 0.5756665468215942,
"learning_rate": 5.421371997298781e-06,
"loss": 0.4057,
"step": 800
},
{
"epoch": 1.3917525773195876,
"grad_norm": 0.5290353298187256,
"learning_rate": 5.156715846877234e-06,
"loss": 0.4018,
"step": 810
},
{
"epoch": 1.40893470790378,
"grad_norm": 0.5541085600852966,
"learning_rate": 4.896419970112499e-06,
"loss": 0.4039,
"step": 820
},
{
"epoch": 1.4261168384879725,
"grad_norm": 0.5141167640686035,
"learning_rate": 4.640718704123472e-06,
"loss": 0.3968,
"step": 830
},
{
"epoch": 1.443298969072165,
"grad_norm": 0.556989312171936,
"learning_rate": 4.389842249629095e-06,
"loss": 0.4049,
"step": 840
},
{
"epoch": 1.4604810996563573,
"grad_norm": 0.5239170789718628,
"learning_rate": 4.144016463705081e-06,
"loss": 0.4046,
"step": 850
},
{
"epoch": 1.47766323024055,
"grad_norm": 0.5183722972869873,
"learning_rate": 3.903462656451074e-06,
"loss": 0.3988,
"step": 860
},
{
"epoch": 1.4948453608247423,
"grad_norm": 0.5335853695869446,
"learning_rate": 3.668397391751336e-06,
"loss": 0.3968,
"step": 870
},
{
"epoch": 1.5120274914089347,
"grad_norm": 0.5347002148628235,
"learning_rate": 3.4390322923083385e-06,
"loss": 0.4013,
"step": 880
},
{
"epoch": 1.529209621993127,
"grad_norm": 0.5166622400283813,
"learning_rate": 3.2155738491247303e-06,
"loss": 0.391,
"step": 890
},
{
"epoch": 1.5463917525773194,
"grad_norm": 0.5376216173171997,
"learning_rate": 2.99822323560525e-06,
"loss": 0.3999,
"step": 900
},
{
"epoch": 1.563573883161512,
"grad_norm": 0.5535770058631897,
"learning_rate": 2.787176126445923e-06,
"loss": 0.4,
"step": 910
},
{
"epoch": 1.5807560137457046,
"grad_norm": 0.5172690153121948,
"learning_rate": 2.582622521473588e-06,
"loss": 0.4084,
"step": 920
},
{
"epoch": 1.597938144329897,
"grad_norm": 0.5224654674530029,
"learning_rate": 2.384746574594343e-06,
"loss": 0.4019,
"step": 930
},
{
"epoch": 1.6151202749140894,
"grad_norm": 0.5353347659111023,
"learning_rate": 2.1937264280049365e-06,
"loss": 0.3987,
"step": 940
},
{
"epoch": 1.6323024054982818,
"grad_norm": 0.5235323905944824,
"learning_rate": 2.0097340518163e-06,
"loss": 0.4012,
"step": 950
},
{
"epoch": 1.6494845360824741,
"grad_norm": 0.5088032484054565,
"learning_rate": 1.8329350892336616e-06,
"loss": 0.3925,
"step": 960
},
{
"epoch": 1.6666666666666665,
"grad_norm": 0.5251315236091614,
"learning_rate": 1.6634887074325844e-06,
"loss": 0.4053,
"step": 970
},
{
"epoch": 1.6838487972508591,
"grad_norm": 0.5620279908180237,
"learning_rate": 1.5015474542651964e-06,
"loss": 0.4015,
"step": 980
},
{
"epoch": 1.7010309278350515,
"grad_norm": 0.5570114254951477,
"learning_rate": 1.3472571209256157e-06,
"loss": 0.404,
"step": 990
},
{
"epoch": 1.718213058419244,
"grad_norm": 0.5131509900093079,
"learning_rate": 1.200756610698205e-06,
"loss": 0.3879,
"step": 1000
},
{
"epoch": 1.718213058419244,
"eval_loss": 0.47505149245262146,
"eval_runtime": 21.8169,
"eval_samples_per_second": 4.95,
"eval_steps_per_second": 2.475,
"step": 1000
},
{
"epoch": 1.7353951890034365,
"grad_norm": 0.5027723908424377,
"learning_rate": 1.0621778139068128e-06,
"loss": 0.3909,
"step": 1010
},
{
"epoch": 1.7525773195876289,
"grad_norm": 0.5319671034812927,
"learning_rate": 9.316454891775983e-07,
"loss": 0.3951,
"step": 1020
},
{
"epoch": 1.7697594501718212,
"grad_norm": 0.5133644938468933,
"learning_rate": 8.092771511223185e-07,
"loss": 0.3917,
"step": 1030
},
{
"epoch": 1.7869415807560136,
"grad_norm": 0.502972424030304,
"learning_rate": 6.951829645431985e-07,
"loss": 0.3925,
"step": 1040
},
{
"epoch": 1.8041237113402062,
"grad_norm": 0.521693766117096,
"learning_rate": 5.894656452546343e-07,
"loss": 0.3984,
"step": 1050
},
{
"epoch": 1.8213058419243986,
"grad_norm": 0.48463624715805054,
"learning_rate": 4.92220367611006e-07,
"loss": 0.3977,
"step": 1060
},
{
"epoch": 1.8384879725085912,
"grad_norm": 0.5170854926109314,
"learning_rate": 4.035346788238681e-07,
"loss": 0.402,
"step": 1070
},
{
"epoch": 1.8556701030927836,
"grad_norm": 0.5153931975364685,
"learning_rate": 3.2348842014562564e-07,
"loss": 0.397,
"step": 1080
},
{
"epoch": 1.872852233676976,
"grad_norm": 0.5183135271072388,
"learning_rate": 2.5215365499069446e-07,
"loss": 0.4038,
"step": 1090
},
{
"epoch": 1.8900343642611683,
"grad_norm": 0.5210497379302979,
"learning_rate": 1.895946040588048e-07,
"loss": 0.4018,
"step": 1100
},
{
"epoch": 1.9072164948453607,
"grad_norm": 0.51735520362854,
"learning_rate": 1.3586758751890638e-07,
"loss": 0.3945,
"step": 1110
},
{
"epoch": 1.9243986254295533,
"grad_norm": 0.5072469711303711,
"learning_rate": 9.10209743056889e-08,
"loss": 0.411,
"step": 1120
},
{
"epoch": 1.9415807560137457,
"grad_norm": 0.5099995136260986,
"learning_rate": 5.509513857437365e-08,
"loss": 0.4021,
"step": 1130
},
{
"epoch": 1.9587628865979383,
"grad_norm": 0.5173975229263306,
"learning_rate": 2.812242335299642e-08,
"loss": 0.3981,
"step": 1140
},
{
"epoch": 1.9759450171821307,
"grad_norm": 0.5263886451721191,
"learning_rate": 1.0127111424872437e-08,
"loss": 0.3999,
"step": 1150
},
{
"epoch": 1.993127147766323,
"grad_norm": 0.5348356366157532,
"learning_rate": 1.125403467490127e-09,
"loss": 0.3927,
"step": 1160
},
{
"epoch": 2.0,
"step": 1164,
"total_flos": 2.8348028555949507e+18,
"train_loss": 0.47038160638301235,
"train_runtime": 14726.6059,
"train_samples_per_second": 1.265,
"train_steps_per_second": 0.079
}
],
"logging_steps": 10,
"max_steps": 1164,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.8348028555949507e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}