OpsEval / data_v2 /network_en_qa_gen.csv
Junetheriver's picture
update 05-15
32e04fa
raw
history blame
556 Bytes
name,rouge1,rouge2,rouge_l,rouge_lsum,score,bp,sys_len,gpt4_score
GPT-3.5-turbo,13.38,5.65,12.13,12.26,6.78,1,2966,8.47
LLaMA-2-70B,8.69,2.51,7.62,7.74,4.2,1,4970,7.28
LLaMA-2-13B,5.75,1.68,5.03,4.98,3.43,1,8239,7.16
Chinese-Alpaca-2-13B,3.48,0.96,3.19,3.25,1.85,1,14716,6.66
Baichuan-13B-Chat,5.58,1.85,4.66,4.76,0.35,1,9577,5.85
Qwen-7B-Chat,13.03,4.76,11.61,11.82,4.33,1,3091,5.63
ChatGLM2-6B,10.43,3.24,9.82,9.71,5.07,0.91,2492,4.88
InternLM-7B,14.34,5.39,13.3,13.27,0.54,1,3112,4.52
Chinese-LLaMA-2-13B,9.18,2.9,9.22,9.19,0.24,1,32006,2.39