Dataset
Hi team , this seems to be a very great work in legal domain. Glad that you have open sourced the model. Is there any idea to open source the dataset used for continued pretraining and dataset used for IFT (General and Legal instructions)
Thanks a lot for the kind words.
We are iterating internally and having some discussions! We will keep you posted!
@PierreColombo Could you provide a time frame approximately on when the dataset used for the base and Instruction fine tuned model will be available ?
Hi @PierreColombo , any update on this?
I have an experimented with this one yet but why not just change the settings in the config file to adjust the model to perform to its needs because you can make a model more introspective on things or more philosophical was it answers at least it seems like it based on the config file you just got to find which corresponding numbers match which which internal architecture.
What kind of update are you trying to do this llm I would like to know I would like to use this llm in experiment I'd like to extract its internal architecture for an experiment,