Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Conversation Evaluation on Proprietary Chinese 500 Multi-turn Dialogues (test)
Loading...
79.4
Win Rate vs GPT4
C-SFT-Empathy
7.432
26.116
44.8
63.484
Sep 10, 2024
Win Rate vs GPT4
Win Rate vs Llama3-70B-Instruct
Chinese Score (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Win Rate vs GPT4
Win Rate vs Llama3-70B-Instruct
Chinese Score (%)
C-SFT-Empathy
Parameters=70B
2024.09
79.4
97.5
77.3
Llama3-70B-Instruct
Parameters=70B
2024.09
10.2
-
9
Feedback
Search any
task
Search any
task