CapyberaHermesYi-34B-ChatML-200K

Model Note

This model has had the missing chat template added. I have only very briefly tested it to ensure it can talk without errors. I wanted to save it to my repo before the pod blew up, imploded, or got sucked into an alternative dimension - because RunPod. I will be using this model for further testing and hopefully (endless trials) perhaps fine-tuning.

I am fixing up older models for their ability to be better co-creative partners than modern guardrailed slop that has eight different coffee makers and an air fryer but sounds like corporate anus, and the stank just doesn't wash out of it.

When I get to test this a bit further, I will update.

Please visit NousResearch/Nous-Capybara-34B for details on this model and to see all the great models Nous makes.

Nous-Capybara-34B V1.9

This is trained on the Yi-34B model with 200K context length, for 3 epochs on the Capybara dataset!

First 34B Nous model and first 200K context length Nous model!

The Capybara series is the first Nous collection of models made by fine-tuning mostly on data created by Nous in-house.

We leverage our novel data synthesis technique called Amplify-instruct (Paper coming soon), the seed distribution and synthesis method are comprised of a synergistic combination of top performing existing data synthesis techniques and distributions used for SOTA models such as Airoboros, Evol-Instruct(WizardLM), Orca, Vicuna, Know_Logic, Lamini, FLASK and others, all into one lean holistically formed methodology for the dataset and model. The seed instructions used for the start of synthesized conversations are largely based on highly regarded datasets like Airoboros, Know logic, EverythingLM, GPTeacher and even entirely new seed instructions derived from posts on the website LessWrong, as well as being supplemented with certain in-house multi-turn datasets like Dove(A successor to Puffin).

While performing great in it's current state, the current dataset used for fine-tuning is entirely contained within 20K training examples, this is 10 times smaller than many similar performing current models, this is signficant when it comes to scaling implications for our next generation of models once we scale our novel syntheiss methods to significantly more examples.

Process of creation and special thank yous!

This model was fine-tuned by Nous Research as part of the Capybara/Amplify-Instruct project led by Luigi D.(LDJ) (Paper coming soon), as well as significant dataset formation contributions by J-Supha and general compute and experimentation management by Jeffrey Q. during ablations.

Special thank you to A16Z for sponsoring our training, as well as Yield Protocol for their support in financially sponsoring resources during the R&D of this project.