Fine tune Mistral-7B (or other model <=7B e.g CodeLlama) for SQL generation on 24GB GPU
Primary LanguageJupyter Notebook