Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
Python cliconsumer-gpudpofine-tuningggufhuggingfacellmllmopslocal-ailocal-llmloralow-vrammachine-learningollamapeftpythonpytorchqlorasfttransformers
"""Global constants and defaults."""
APP_NAME = "soup"
CONFIG_FILE = "soup.yaml"
SOUP_DIR = ".soup"
EXPERIMENTS_DB = "experiments.db"
GITHUB_URL = "https://github.com/MakazhanAlpamys/Soup"
DEFAULT_CHAT_TEMPLATE = (
"{% for message in messages %}"
"{% if message['role'] == 'system' %}"
"{{ message['content'] + '\\n' }}"
"{% elif message['role'] == 'user' %}"
"{{ 'User: ' + message['content'] + '\\n' }}"
"{% elif message['role'] == 'assistant' %}"
"{{ 'Assistant: ' + message['content'] + '\\n' }}"
"{% endif %}{% endfor %}"
)