4 ms·
Based: An Educational and Effective Sequence Mixer
- pama 3y agoThis blog post describes “Based” a mamba-adjacent architecture with huge context that achieves similarly low perplexity in NLP. The request came up in: https://news.ycombinator.com/item?id=38717236 https://news.ycombinator.com/item?id=38717236