WizardLM
WizardLM copied to clipboard
nlpxucan
Family of instruction-following LLMs powered by Evol-Instruct: WizardLM, WizardCoder
I use these LLM's for creative writing and with Vicuna I'm able to get it to revise sections, like add or remove details, make things longer or shorter, etc (with...
When I try to load WizardLM into my CMD it says "Loading wizardLM-7B-GPTQ-4bit-128g... Can't determine model type from model name. Please specify it manually using --model_type argument" and then closes....
It looks like you are doing a full fine-tuning run, have you experimented with LoRA?
wizardLM can't receive chunk of text and rewrite it as specified. I use simple prompt template. ``` TEXT: {book} DIRECTIVE: read the text and formulate N relevant questions with answers....
Firstly, thanks for publicly releasing the instruction dataset. While looking through the [dataset](https://huggingface.co/datasets/victor123/evol_instruct_70k), I've noticed several examples where the `output` is already present in the `instruction` with barely any changes....
All modern transformer models has problems with math. that's because they tokenise numbers weirdly. Your model supposed to be better in math than others. But main reason, why are transformers...
Your README claims this model outperforms ChatGPT with "complex tasks." How should we define complex tasks? I asked it to write me shader code for Open GL ES 2.0, and...
Hi authors, thank you for releasing code and data for this project. I am confused about the following part in the paper. > For fair comparison, we replace Alpaca’s original...
