llama2.c
llama2.c copied to clipboard
About runq.c
I follow pesvut's method #395
use"curl -L -O https://huggingface.co/karpathy/tinyllamas/resolve/main/stories15M.pt
python export.py tinyllama-15m --checkpoint ./stories15M.pt --version -1 "
Convert "stories15M.pt" to ”config.json“ and "pytorch_model.bin"
then use "python export.py tinyllama-15m.bin -hf .\tinyllama-15m" and "python export.py tinyllama-8bit.bin --version 2 --hf .\tinyllama-15m" Convert "pytorch_model.bin" to ”tinyllama-15m.bin“ and "tinyllama-8bit.bin"
Normal generation when executing the run program.But when using the runq program, there is no output, and after waiting for a while, the program ends
I hope someone can help me answer the question of which step I quantify that went wrong