logical-reasoning / data /Qwen2.5-7B-Instruct_shots_metrics.csv
dh-mc's picture
clean up data
0e5f859
raw
history blame
947 Bytes
shots,model,run,accuracy,precision,recall,f1,ratio_valid_classifications
0,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-00,0.6436666666666667,0.717651042027604,0.6436666666666667,0.6066932578767255,1.0
5,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-05,0.63,0.7622571683877091,0.63,0.6151126410759672,0.998
10,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-10,0.677,0.7663956674673086,0.677,0.6770580664953397,0.9796666666666667
20,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-20,0.7343333333333333,0.7730863408305184,0.7343333333333333,0.7243291573141537,0.807
30,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-30,0.765,0.7840432806350224,0.765,0.7512220322751986,0.805
40,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-40,0.757,0.7733827213068922,0.757,0.7427592763321033,0.8546666666666667
50,Qwen2.5-7B-Instruct,Qwen/Qwen2.5-7B-Instruct/shots-50,0.758,0.763149679724481,0.758,0.7376580515312735,0.7563333333333333