3 ms·

Completely the opposite. - it is NOT backed directly or owned by any VC funded company - it is 100% OSS driven by the community (Apache 2 license) - it’s cur

Completely the opposite. - it is NOT backed directly or owned by any VC funded company - it is 100% OSS driven by the community (Apache 2 license) - it’s currently the top OSS chat model that can be used commercially on the chatbot arena score board - IMO it is undertrained, so expanding the training data alone will make it much better (however for the sake of this paper, we wanted to focus on architecture not training data, so we compared similarly trained models) And yes we do have multiple experiments and plans to make it better. It’s a list, and we will not know which is final until we try. Individual members can go to great lengths on what they are working on For better or worse, being truly OSS means our initiatives are more disorganized then a centrally planned org

bluepoint 3y ago
Can beginners without money with by PhDs, contribute? If so what would be the best way to start?
- ftxbro 3y ago
  i would guess, go to their discord. also they put their github so you could fix a bug in their github.
- pico_creator 3y ago
  There are lots of really low hanging fruits - integrating this with AI platform X/Y/Z - setting up evals - improving the code quality - making a how to guide (it’s stuck on my todo list) - helping with dataset - doing silly experiments on how the architecture work (and if the changes give good result) - etc etc One of the community goals is to make this a model for EVERYONE on earth that means we need quality dataset for all the non English languages So even on that level there are things to do ( find something that interest you on the community )
redox99 3y ago
> it’s currently the top OSS chat model that can be used commercially on the chatbot arena score board To be fair, that filters the majority of models in the scoreboard.
- pico_creator 3y ago
  Dun we all wish this wasn’t the case? Where we have more OSS models to choose from without weird rule lawyering gotchas. Or needing to be from a research institute / a license to download the weights

bluepoint 3y ago

Can beginners without money with by PhDs, contribute? If so what would be the best way to start?

ftxbro 3y ago
i would guess, go to their discord. also they put their github so you could fix a bug in their github.
pico_creator 3y ago
There are lots of really low hanging fruits - integrating this with AI platform X/Y/Z - setting up evals - improving the code quality - making a how to guide (it’s stuck on my todo list) - helping with dataset - doing silly experiments on how the architecture work (and if the changes give good result) - etc etc One of the community goals is to make this a model for EVERYONE on earth that means we need quality dataset for all the non English languages So even on that level there are things to do ( find something that interest you on the community )

redox99 3y ago

> it’s currently the top OSS chat model that can be used commercially on the chatbot arena score board To be fair, that filters the majority of models in the scoreboard.

pico_creator 3y ago
Dun we all wish this wasn’t the case? Where we have more OSS models to choose from without weird rule lawyering gotchas. Or needing to be from a research institute / a license to download the weights