初始化项目，由ModelHub XC社区提供模型

Model: bigscience/bloomz-7b1-p3
Source: Original Platform

evaluation_bloomz-7b1-p3/evaluation_xnliht/xnli/merged.csv

@@ -0,0 +1,50 @@
+dataset,prompt,metric,value
+xnli_ar,GPT-3 style_arht,accuracy,0.3634538152610442
+xnli_ar,MNLI crowdsource_arht,accuracy,0.3437751004016064
+xnli_ar,can we infer_arht,accuracy,0.40923694779116465
+xnli_ar,guaranteed/possible/impossible_arht,accuracy,0.342570281124498
+xnli_ar,justified in saying_arht,accuracy,0.4321285140562249
+xnli_ar,median,accuracy,0.3634538152610442
+xnli_es,GPT-3 style_esht,accuracy,0.4795180722891566
+xnli_es,MNLI crowdsource_esht,accuracy,0.3333333333333333
+xnli_es,can we infer_esht,accuracy,0.3333333333333333
+xnli_es,guaranteed/possible/impossible_esht,accuracy,0.529718875502008
+xnli_es,justified in saying_esht,accuracy,0.3333333333333333
+xnli_es,median,accuracy,0.3333333333333333
+xnli_fr,GPT-3 style_frht,accuracy,0.45863453815261046
+xnli_fr,MNLI crowdsource_frht,accuracy,0.42730923694779116
+xnli_fr,can we infer_frht,accuracy,0.40963855421686746
+xnli_fr,guaranteed/possible/impossible_frht,accuracy,0.37309236947791163
+xnli_fr,justified in saying_frht,accuracy,0.4710843373493976
+xnli_fr,median,accuracy,0.42730923694779116
+xnli_hi,GPT-3 style_hiht,accuracy,0.3542168674698795
+xnli_hi,MNLI crowdsource_hiht,accuracy,0.4389558232931727
+xnli_hi,can we infer_hiht,accuracy,0.41566265060240964
+xnli_hi,guaranteed/possible/impossible_hiht,accuracy,0.4927710843373494
+xnli_hi,justified in saying_hiht,accuracy,0.4562248995983936
+xnli_hi,median,accuracy,0.4389558232931727
+xnli_sw,GPT-3 style_swht,accuracy,0.3389558232931727
+xnli_sw,MNLI crowdsource_swht,accuracy,0.3257028112449799
+xnli_sw,can we infer_swht,accuracy,0.3429718875502008
+xnli_sw,guaranteed/possible/impossible_swht,accuracy,0.3718875502008032
+xnli_sw,justified in saying_swht,accuracy,0.3409638554216867
+xnli_sw,median,accuracy,0.3409638554216867
+xnli_ur,GPT-3 style_urht,accuracy,0.3646586345381526
+xnli_ur,MNLI crowdsource_urht,accuracy,0.3538152610441767
+xnli_ur,can we infer_urht,accuracy,0.3610441767068273
+xnli_ur,guaranteed/possible/impossible_urht,accuracy,0.37670682730923694
+xnli_ur,justified in saying_urht,accuracy,0.3377510040160643
+xnli_ur,median,accuracy,0.3610441767068273
+xnli_vi,GPT-3 style_viht,accuracy,0.3357429718875502
+xnli_vi,MNLI crowdsource_viht,accuracy,0.3477911646586345
+xnli_vi,can we infer_viht,accuracy,0.3333333333333333
+xnli_vi,guaranteed/possible/impossible_viht,accuracy,0.39759036144578314
+xnli_vi,justified in saying_viht,accuracy,0.3333333333333333
+xnli_vi,median,accuracy,0.3357429718875502
+xnli_zh,GPT-3 style_zhht,accuracy,0.348995983935743
+xnli_zh,MNLI crowdsource_zhht,accuracy,0.45100401606425705
+xnli_zh,can we infer_zhht,accuracy,0.3931726907630522
+xnli_zh,guaranteed/possible/impossible_zhht,accuracy,0.39879518072289155
+xnli_zh,justified in saying_zhht,accuracy,0.3827309236947791
+xnli_zh,median,accuracy,0.3931726907630522
+multiple,average,multiple,0.3742469879518072