notahuman

3 ms·

1. 1.▲
   by notahuman 4y ago

   Recently, I conducted a few fine-tuning experiments on a multitask instruction dataset on 1b7 LLM, primarily related to open-ended long story generation. However, it was observed that after generating nearly 300 tokens, the quality of the g
2. 2.▲
   Correlation between the model scale and the quality of long text generation?(old.reddit.com)

   1 points by notahuman 4y ago|1 comments
3. 3.▲
   Untrue Statements in “Real Differences Between OT and CRDT for Co-Editors”

   3 points by notahuman 8y ago|0 comments
4. 4.▲
   by notahuman 12y ago

   Wow, honestly i want a similar life just like you.
5. 5.▲
   The most simple VPN app ever (Canada for now)(vpndefender.com)

   2 points by notahuman 12y ago|0 comments