4o
Collection
58 items • Updated
• 2
YAML Metadata Warning: empty or missing yaml metadata in repo card
Check out the documentation for more information.
Hyper-Creative Grok V1
Overview
Expert-tier literary generation using an expanded author seed (17 authors). Features heavy stylistic mirroring and narrative state tracking [STATE] tags.
Dataset Details
- Total Samples: [Processed and Cleaned]
- Format: ShareGPT-compatible JSONL
- Primary Category: Creative Writing / Stylistic Mirroring
- Model Used: x-ai/grok-4.1-fast (OpenRouter)
Features
- Strict Deduplication: All duplicate generations removed.
- Chronological Sorting: Sorted by generation timestamp.
- Narrative Coherence: High token-density samples featuring internal world-state tracking and consistent character knowledge.
License
License-free / Creative Commons (Open Data)
Stats
| Metric | Value |
|---|---|
| Total prompt tokens | 2,430,791 |
| Total completion tokens | 5,646,369 |
| Total tokens | 8,077,160 |
| Total cost | $91.99 (USD) |
| Average turns | 1.03 |
| Average tool calls | 0.00 |
| Average tokens per row | 2,377.73 |
Cost estimated using Grok pricing on OpenRouter ($3.0/M input, $15.0/M output)
- Downloads last month
- 46