eas / nous-capybara

Nous-Capybara-34B V1.9 in select quantizations

34B

795 Pulls Updated 9 months ago

10 months ago

b977635d75a7 · 21GB

model
llama
·
34.4B
·
Q4_K_M
template
USER: {{ .Prompt }} ASSISTANT:
params
{"num_ctx":10240,"stop":["</s>"]}

Readme

Nous-Capybara-34B V1.9

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

I’m providing q4_K_M & q3_K_M quantizations. Models are set up for 10k context. You can try larger contexts with >>>/set parameter num_ctx XXXXXXX in the CLI, or by creating a custom model.

NousResearch/Nous-Capybara-34B on Hugging Face