Model Card for Extended-Mind-MPT-7b

Extended Mind MPT-7b, as described in Supersizing Transformers.

Model Description

This model implements active externalism for MPT's 7b model. The model weights have not been edited. Original architecture and code by Mosaic ML.

For more details on active externalism, check out our blog!

• Developed by: Normal Computing, Adapted from Mosacic ML
• License: Apache 2.0

Limitations

This model is part of ongoing research at Normal Computing.