| # GPT4ALL or llama-cpp-python model_kwargs | |
| # GPT4ALl GPT-J type, from model explorer choice, so downloads | |
| model_name_gptj=ggml-gpt4all-j-v1.3-groovy.bin | |
| # llama-cpp-python type, supporting version 3 quantization, here from locally built llama.cpp q4 v3 quantization | |
| # below uses prompt_type=wizard2 | |
| model_path_llama=WizardLM-7B-uncensored.ggmlv3.q8_0.bin | |
| # below assumes max_new_tokens=256 | |
| n_ctx=1792 | |
| # uncomment below if using llama-cpp-pyton with cublas built in | |
| # n_gpu_layers=20 | |
| # GPT4ALl LLaMa type, supporting version 2 quantization, here from model explorer choice so downloads | |
| model_name_gpt4all_llama=ggml-wizardLM-7B.q4_2.bin | |
| # PDF_CLASS_NAME=UnstructuredPDFLoader | |