this is a very early pre-release model. see the github page for more details https://github.com/fakerybakery/pruned-yi

do not expect high-quality generations. longer prompts work better. trained on a mix of cosmopedia + chinese wikipedia (mix in a 2:1 ratio to preserve bilingual abilities). licensed under apache 2.0.

please open a github issue for any questions

Downloads last month: 324

Safetensors

Model size

3.29B params

Tensor type

BF16

Text Generation

This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.