Tiny-Aya-Base Blind Spots Evaluation Dataset
Domain:
natural language processingRecord type:
dataset
Creator:
ANI
Host:
CohereLabs/tiny-aya-base
Architecture: Cohere2 (Cohere Command R family)
Parameters: ~3B
Type: Raw pre-trained base model — not instruction-tuned or RLHF'd
Languages: 70+ languages, with emphasis on low-resource language coverage
Context Length: 8K tokens
Released: February 2025