| ▲ | Model can accept 1M tokens doesn't mean it can reason across those 1M tokens(twitter.com) | |
| 2 points by annjose 13 hours ago | 1 comments | ||
| ▲ | turtleyacht 13 hours ago | parent [-] | |
Looks like there is a memory hierarchy [1] even in LLMs. | ||
| ▲ | Model can accept 1M tokens doesn't mean it can reason across those 1M tokens(twitter.com) | |
| 2 points by annjose 13 hours ago | 1 comments | ||
| ▲ | turtleyacht 13 hours ago | parent [-] | |
Looks like there is a memory hierarchy [1] even in LLMs. | ||