search

CERC-AAI / multimodal

An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.
Apache License 2.0
8 stars 3 forks source link

Lora locations #45

Closed daniel-z-kaplan closed 1 year ago

daniel-z-kaplan commented 1 year ago

@kshitijkg

daniel-z-kaplan commented 1 year ago

The only files modified are megatron/training.py, and megatron/models/adapter.py