hynky's picture
hynky HF staff
new plotting code (JIT)
0932e7b
raw
history blame
1.28 kB
{"data":{"ind_minhash-CC-MAIN-2019-18":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3308933284133672,0.35846718959510326,0.37850185949355364,0.39194786734879017,0.39600365422666073,0.40486439503729343,0.4064061753451824,0.41104014590382576,0.41393135115504265,0.41698802448809147,0.42121383734047413,0.4219294786453247,0.4234823901206255,0.42346264235675335,0.42699199728667736],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3308620806783438,0.3493095366284251,0.367914117872715,0.37658837065100664,0.3858313206583261,0.3908915650099516,0.3968510050326586,0.39992102794349194,0.40259181894361973,0.4055726025253534,0.4074157159775495,0.40804907679557795,0.40894216299057007,0.4123705606907606,0.4108315110206604],"label":"WET data"}},"layout":{"xaxis":{"title":{"text":"Training tokens (billions)"}},"title":{"text":"WET data is worse than data extracted from WARC"}}}