Instruction Following Evaluation on Instruction Tuning with GPT-4
71.29ScoreClaude3
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Claude3evaluation_scale=percentage-based2024.09 | 71.29 | — | — | |
| Qwenevaluation_scale=percentage-based2024.09 | 70.14 | — | — | |
| GLM-4evaluation_scale=percentage-based2024.09 | 69.89 | — | — | |
| GPT-4 Turboevaluation_scale=percentage-based2024.09 | 69.25 | — | — | |
| GPT-4evaluation_scale=percentage-based2024.09 | 67.58 | — | — | |
| Aggregate Statistics2024.09 | — | 69.63 | 1.49 |