Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -82,7 +82,7 @@ The subsets can be for this training step are derived from 2 different sources:
|
|
| 82 |
### Methodology: Reasoning-DE & Reasoning-EN
|
| 83 |
- Providing Persona Descriptions and rewriting in a similar style with a different focus area and name in german/english language
|
| 84 |
- Generating Simple Logical Problems out of Persona-specific Views & Language.
|
| 85 |
-
- Generating Approaches, Thinking-Steps & Solutions separately verified by Llama-3.1-
|
| 86 |
- Quality assurance and validation
|
| 87 |
|
| 88 |
## Task Types and Examples
|
|
|
|
| 82 |
### Methodology: Reasoning-DE & Reasoning-EN
|
| 83 |
- Providing Persona Descriptions and rewriting in a similar style with a different focus area and name in german/english language
|
| 84 |
- Generating Simple Logical Problems out of Persona-specific Views & Language.
|
| 85 |
+
- Generating Approaches, Thinking-Steps & Solutions separately verified by Llama-3.1-405B-Instruct
|
| 86 |
- Quality assurance and validation
|
| 87 |
|
| 88 |
## Task Types and Examples
|