pyllama
pyllama copied to clipboard
juncongmoo
LLaMA: Open and Efficient Foundation Language Models
Hello, I use 4 V100 GPU, and I load the 30B model, I want to modify the example.py code to input my promths. But it doesnot work. My code as...
We evaluate llama using 100 examples of the [`SQuAD`](https://huggingface.co/datasets/squad) dataset with the [Open-evals](https://github.com/open-evals/evals) framework, which extends OpenAI's Evals for different language models. We consider the sentence immediately following the prompt...
I have cloded the repo, installed all requirements including the CMake and itree based on one of the reported issues, still I run into the following traceback error when trying...
``` ~/GPT/pyllama_data/pyllama$ python inference.py --ckpt_dir $CKPT_DIR --tokenizer_path $TOKENIZER_PATH Traceback (most recent call last): File "/home/ildar/GPT/pyllama_data/pyllama/inference.py", line 82, in run( File "/home/ildar/GPT/pyllama_data/pyllama/inference.py", line 50, in run generator = load( File "/home/ildar/GPT/pyllama_data/pyllama/inference.py",...
I am new to AI and trying to use `llama2` model locally using `pyllama`. I tried different options, but nothing seems to work. I downloaded llama using https://github.com/facebookresearch/llama. Here is...
My card has 12GB of RAM, that's not a case covered anywhere i could see. Would this allow me to do more (run the larger models, etc)? Any chances to...
It stucks at downloading file for 3 hours but still unfinished: ``` (llamachat) [ec2-user@ip-172-31-6-66 llamachat]$ python -m llama.download --model_size 7B --folder ./pyllama_data/ ❤️ Resume download is supported. You can ctrl-c...
A bit confused here. In README.md, users are asked to donwload LLaMA model files first. Then quantize examples use **decapoda-research/llama-7b-hf**. How to quantize the downloaded LLaMA model files(for example, consolidated.00.pth...
python webapp_single.py --ckpt_dir $CKPT_DIR --tokenizer_path $TOKENIZER_PATH when i run this command, these is a error : no module named llama??
Metadata
Owner
Metadata
LLaMA: Open and Efficient Foundation Language Models