OxMini-v2 / base_training_log.csv
Shivam3002's picture
Publish measured OxMini v2 replay-SFT release
84bf611 verified
Raw
History Blame Contribute Delete
5.73 kB
step,train_loss,val_loss,learning_rate,tokens_per_second,elapsed_seconds
0,4.83918947,4.84112531,0.0,0.0,0.0
10,4.75729237,,0.0000500000,823.59,31.083
20,4.33398438,,0.0001000000,899.71,59.537
30,3.87951114,,0.0001500000,890.71,88.279
40,3.55222142,,0.0002000000,882.17,117.299
50,3.26539285,,0.0002500000,858.64,147.114
60,2.95845702,,0.0003000000,899.41,175.578
70,2.69584179,,0.0002999585,893.10,204.242
80,2.48971870,,0.0002998150,871.94,233.603
90,2.32306039,,0.0002995691,886.01,262.497
100,2.18203435,2.11185485,0.0002992211,878.31,291.644
110,2.09628744,,0.0002987711,875.84,322.643
120,2.00068557,,0.0002982195,892.25,351.335
130,1.95019857,,0.0002975668,881.19,380.387
140,1.90446488,,0.0002968134,925.32,408.054
150,1.84049996,,0.0002959599,927.69,435.650
160,1.77750546,,0.0002950069,902.67,464.011
170,1.78323630,,0.0002939552,888.26,492.832
180,1.72335373,,0.0002928056,880.71,521.901
190,1.68470315,,0.0002915590,890.63,550.645
200,1.69683789,1.64557241,0.0002902162,881.68,579.682
210,1.64066246,,0.0002887784,876.12,610.665
220,1.61724705,,0.0002872465,870.39,640.078
230,1.58990306,,0.0002856218,879.51,669.185
240,1.56097727,,0.0002839056,872.79,698.517
250,1.56202164,,0.0002820990,885.68,727.422
260,1.55520117,,0.0002802035,858.37,757.246
270,1.51839741,,0.0002782205,864.63,786.855
280,1.52199612,,0.0002761516,902.50,815.222
290,1.49851342,,0.0002739982,900.47,843.652
300,1.46621981,1.48078024,0.0002717621,889.49,872.434
310,1.50000474,,0.0002694449,886.51,903.109
320,1.44315324,,0.0002670483,856.73,932.990
330,1.47954578,,0.0002645743,870.81,962.388
340,1.42425752,,0.0002620247,862.96,992.054
350,1.43625284,,0.0002594014,872.66,1021.390
360,1.45359471,,0.0002567064,881.50,1050.433
370,1.42159264,,0.0002539417,897.46,1078.958
380,1.39923534,,0.0002511096,877.63,1108.128
390,1.42354652,,0.0002482120,911.89,1136.202
400,1.39415280,1.39577638,0.0002452512,879.48,1165.310
410,1.38192692,,0.0002422295,891.49,1195.802
420,1.37066246,,0.0002391492,906.57,1224.041
430,1.42119601,,0.0002360125,892.36,1252.730
440,1.35556067,,0.0002328219,884.30,1281.679
450,1.35521977,,0.0002295798,895.32,1310.273
460,1.35225177,,0.0002262887,871.77,1339.639
470,1.36019539,,0.0002229510,895.18,1368.237
480,1.33163208,,0.0002195694,893.62,1396.886
490,1.29739975,,0.0002161463,892.58,1425.567
500,1.34938394,1.33227201,0.0002126843,884.76,1454.502
510,1.30087914,,0.0002091861,874.05,1485.561
520,1.31972418,,0.0002056544,871.82,1514.925
530,1.37210366,,0.0002020918,849.01,1545.078
540,1.32015469,,0.0001985011,881.45,1574.122
550,1.30201854,,0.0001948849,891.57,1602.836
560,1.28928034,,0.0001912460,886.41,1631.717
570,1.30795910,,0.0001875871,920.24,1659.536
580,1.30739516,,0.0001839112,883.61,1688.509
590,1.27574106,,0.0001802208,893.08,1717.174
600,1.29050994,1.28732845,0.0001765189,867.79,1746.674
610,1.28455565,,0.0001728083,868.59,1778.005
620,1.31432481,,0.0001690917,888.28,1806.825
630,1.29894265,,0.0001653720,867.22,1836.345
640,1.26539890,,0.0001616521,873.42,1865.656
650,1.27364161,,0.0001579346,884.38,1894.604
660,1.25734029,,0.0001542226,877.41,1923.781
670,1.25216517,,0.0001505187,841.67,1954.197
680,1.21993992,,0.0001468258,865.55,1983.775
690,1.25202012,,0.0001431468,873.88,2013.070
700,1.26427742,1.24851422,0.0001394843,881.43,2042.114
710,1.24795233,,0.0001358412,855.79,2073.889
720,1.23193842,,0.0001322202,869.97,2103.316
730,1.25418483,,0.0001286242,866.40,2132.864
740,1.22694573,,0.0001250557,874.35,2162.144
750,1.23771381,,0.0001215176,880.15,2191.231
760,1.23394274,,0.0001180125,870.66,2220.634
770,1.22236916,,0.0001145431,859.08,2250.433
780,1.23205507,,0.0001111120,901.02,2278.846
790,1.23766434,,0.0001077218,878.18,2307.998
800,1.22899981,1.22271504,0.0001043752,883.32,2336.980
810,1.22301505,,0.0001010745,895.07,2367.402
820,1.22898583,,0.0000978224,859.10,2397.201
830,1.22081829,,0.0000946214,866.10,2426.759
840,1.22254877,,0.0000914737,883.03,2455.750
850,1.23156995,,0.0000883819,868.06,2485.242
860,1.22192689,,0.0000853483,879.73,2514.343
870,1.22415450,,0.0000823752,880.21,2543.427
880,1.20891033,,0.0000794648,868.41,2572.906
890,1.21368941,,0.0000766194,887.01,2601.768
900,1.17150828,1.20737538,0.0000738410,900.59,2630.194
910,1.18077129,,0.0000711320,901.28,2660.386
920,1.20440112,,0.0000684941,879.43,2689.496
930,1.21773757,,0.0000659296,891.89,2718.200
940,1.23081151,,0.0000634403,896.69,2746.749
950,1.17362685,,0.0000610282,882.48,2775.759
960,1.21169600,,0.0000586949,894.33,2804.385
970,1.19922386,,0.0000564425,913.80,2832.401
980,1.19333848,,0.0000542724,875.41,2861.645
990,1.18401357,,0.0000521864,891.57,2890.359
1000,1.20005584,1.18157737,0.0000501861,895.10,2918.960
1010,1.20360632,,0.0000482730,894.53,2949.419
1020,1.19847631,,0.0000464486,879.27,2978.534
1030,1.19095331,,0.0000447141,886.76,3007.404
1040,1.15768230,,0.0000430710,888.97,3036.202
1050,1.19419725,,0.0000415205,884.63,3065.142
1060,1.18620920,,0.0000400638,881.90,3094.171
1070,1.17379215,,0.0000387019,879.99,3123.263
1080,1.16580384,,0.0000374360,839.31,3153.764
1090,1.17508124,,0.0000362669,874.18,3183.050
1100,1.16115761,1.17957374,0.0000351955,881.50,3212.092
1110,1.15750911,,0.0000342228,873.00,3243.264
1120,1.20234993,,0.0000333493,851.11,3273.342
1130,1.18024633,,0.0000325759,830.56,3304.166
1140,1.17465255,,0.0000319030,899.63,3332.623
1150,1.16993039,,0.0000313311,882.38,3361.636
1160,1.20854867,,0.0000308608,862.07,3391.333
1170,1.17070669,,0.0000304923,884.42,3420.279
1180,1.16680235,,0.0000302260,872.11,3449.634
1190,1.22155600,,0.0000300620,874.89,3478.895
1200,1.14508473,1.17385033,0.0000300005,860.65,3508.640