add phi-3-mini eager mode example (#4315)
Summary: This PR adds an example script in eager mode which does inference with/without kv cache enabled. This comes handy when we want to verify that phi-3-mini works in eager mode. Pull Request resolved: https://github.com/pytorch/executorch/pull/4315 Test Plan: ``` python3 -m examples.models.phi-3-mini.eager -s 128 -kv -p "Tell me a story" python3 -m examples.models.phi-3-mini.eager -s 128 -p "Tell me a story" ``` Verify that the model runs faster with kv cache enabled. Reviewed By: JacobSzwejbka Differential Revision: D60061822 Pulled By: helunwencser fbshipit-source-id: 483d2f9e56f9397f78dec805a0c1a110cb1cfc28
H
helunwencser committed
dbf87b0bf35611b2ce701e0fc3f391ee1fcc85fd
Parent: 5b0700b
Committed by Facebook GitHub Bot <facebook-github-bot@users.noreply.github.com>
on 7/25/2024, 6:10:18 PM