Anthropic Helpful Harmless prompts
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| RLHF Backdoor Attack | Anthropic Helpful Harmless prompts (train test) | UHR Rate28.1 | 30 |
Showing 1 of 1 rows
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| RLHF Backdoor Attack | Anthropic Helpful Harmless prompts (train test) | UHR Rate28.1 | 30 |