Patch llama.py for internal build (#7239)
Patch llama.py for internal build (#7239) Summary: As title, minor changes so we can use buck to build.. ``` buck run mode/dev-nosan //executorch/examples/qualcomm/oss_scripts/llama3_2:llama_qnn -- --compile_only --ptq 16a4w --checkpoint /home/chenlai/local/models/consolidated.00.pth --params /home/chenlai/local/models/params.json --tokenizer_model /home/chenlai/local/models/tokenizer.model --prompt "Once" -m SM8650 --model_size 1B --model_mode kv 2>&1 | tee static_llama.log ``` Differential Revision: D66947240
C
cccclai committed
5161d70ac64ae511c68004472c5b5c06cd410efa
Parent: f22d1a3
Committed by GitHub <noreply@github.com>
on 12/10/2024, 12:47:13 AM