huggingface/ratchet

Phi2 Support

Closed this issue 10 months ago · 1 comments

FL33TW00D commented a year ago

Prior Art:
https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/models/phi.py
https://github.com/huggingface/candle/blob/main/candle-transformers/src/models/phi.rs#L40
https://huggingface.co/microsoft/phi-2/blob/main/modeling_phi.py

Missing pieces

RotaryEmbedding - https://github.com/huggingface/candle/blob/main/candle-transformers/src/models/phi.rs#L40
reshape operator

FL33TW00D commented 10 months ago

Completed in #137