Handling long sequences of text seems to challenge the coherence of the model. While the model is trained on long-term text, it struggles to carry information forward indefinitely. The average length of chat observed is 1467 tokens, indicating a limitation compared to today's context windows. The model can go slightly beyond this length but not dramatically. The handling of long sequences is the least proven aspect, which may affect downstream inferences and speculations.

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode