Skip to main content
zenodorestricted

【outputs_encode_to_representations】only two 8B LLMs, only Ir task, augmented responses representations w/o their language materials, w/ original prompt and response representations

<ul> <li>only two 8B LLMs,</li> <li>only Ir task, augmented responses representations w/o their language materials,<br> <ul> <li>12 (valid prompts) ⨉ 2 (LLMs) ⨉ 2 (scores) = 48 files</li> </ul> </li> <li>original prompt and response representations <ul> <li>[1 (prompt) + 1 (response)] ⨉ 2 (LLMs) = 4 files</li> </ul> </li> <li>in total, 48 + 4 = 52 files</li> </ul>

ShareScore

16/100

Overall dataset sharing score

Score breakdown

These five areas show where the dataset supports — or may limit — practical reuse.

Stewardship
4
Harmonization
4
Access
8
Reuse readiness
0
Engagement
0