evaluation_key,benchmark,rows,output_cap,output_tokens,mean_output_tokens,length_capped,length_capped_rate stock_base,milu,3980,8,8638,2.17035175879397,113,0.028391959798994976 stock_base,indicparam,660,8,1332,2.018181818181818,2,0.0030303030303030303 stock_base,indicifeval,602,2048,232636,386.4385382059801,2,0.0033222591362126247 selected_sft,milu,3980,8,8062,2.0256281407035175,17,0.004271356783919598 selected_sft,indicparam,660,8,1320,2.0,0,0.0 selected_sft,indicifeval,602,4096,322238,535.2790697674419,33,0.054817275747508304 sft_plus_gated,milu,3980,8,8044,2.021105527638191,14,0.0035175879396984926 sft_plus_gated,indicparam,660,8,1320,2.0,0,0.0 sft_plus_gated,indicifeval,602,4096,316750,526.1627906976744,32,0.053156146179401995