vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

add mixtral 7bv0.1 q8 lora in gguff format, so we can test perplexity

Browse source
This commit is contained in:
Uzo Nweke 2024-01-20 16:15:47 -05:00
parent da41206bd6
commit 2b137c54bd
1 changed files with 5 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

9
ci/run.sh
View file

@ -640,11 +640,12 @@ function gg_run_mistral_7b_v0_1 {
lora_shakespeare="${path_lora}/ggml-adapter-model.bin"
# TODO - we need mistral lora
gg_wget ${path_lora} https://huggingface.co/slaren/open_llama_7b_v2_shakespeare_lora/resolve/main/adapter_config.json
gg_wget ${path_lora} https://huggingface.co/slaren/open_llama_7b_v2_shakespeare_lora/resolve/main/adapter_model.bin
gg_wget ${path_shakespeare} https://huggingface.co/slaren/open_llama_7b_v2_shakespeare_lora/resolve/main/shakespeare.txt
python3 ../convert-lora-to-ggml.py ${path_lora}
gg_wget ${path_lora} https://huggingface.co/datasets/segmond/mistral_7b_v0_1_q8_0_shakespeare_lora/blob/main/adapter_model.bin
gg_wget ${path_shakespeare} https://huggingface.co/datasets/segmond/mistral_7b_v0_1_q8_0_shakespeare_lora/blob/main/shakespeare.txt
# model is in gguf format since we finetuned from gguf
#python3 ../convert-lora-to-ggml.py ${path_lora}
# f16
(time ./bin/perplexity --model ${model_f16} -f ${shakespeare} -t 1 -ngl 999 -c 2048 -b 512 --chunks 3 ) 2>&1 | tee -a $OUT/${ci}-ppl-shakespeare-f16.log