phi-1 / configuration_mixformer_sequential.py

Commit History

Improves type hinting on configuration arguments.
8d2c4ce

gugarosa commited on

Fixes flash-attn import with a try/except statement
9ed5987

gugarosa commited on

Adds support for MQA/GQA and attention mask during training / fine-tuning.
371fd51

gugarosa commited on

Support for `attention_mask` in forward pass.
d22f35e

gugarosa commited on

Upload MixFormerSequentialForCausalLM
0f4ae0e

suriyagunasekar commited on