logical-reasoning / data /Qwen2.5-1.5B-Instruct-float16_shots_metrics.csv
dh-mc's picture
final few-shots
84958e4
shots,model,run,accuracy,precision,recall,f1,ratio_valid_classifications
0,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-00,0.20166666666666666,0.5269756683734005,0.20166666666666666,0.24069835329504388,0.9223333333333333
5,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-05,0.3933333333333333,0.578886379886985,0.3933333333333333,0.43554636943558694,0.8283333333333334
10,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-10,0.407,0.5820145311822223,0.407,0.459589777544246,0.9156666666666666
20,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-20,0.232,0.5282610881631451,0.232,0.3093707499897376,0.676
30,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-30,0.23,0.5479545947886839,0.23,0.3064381040560128,0.661
40,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-40,0.29233333333333333,0.5608411738006117,0.29233333333333333,0.3751714671158081,0.5206666666666667
50,Qwen2.5-1.5B-Instruct,Qwen/Qwen2.5-1.5B-Instruct/shots-50,0.29,0.5646814860840066,0.29,0.36883826526592467,0.4603333333333333