KO
|
EN
gitlite — search
Search
#javascript
#python
#hacktoberfest
#react
#ai
#typescript
#llm
#go
#golang
#android
#machine-learning
#rust
#deep-learning
#linux
nemesis
★ 63
Open GitHub ↗
Reward Model framework for LLM RLHF
Download README (.md)
Explore Similar Repositories
Relational-NEM
:
Code for the "Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions" paper.
nemanjam.github.io
:
Developer blog website built with Astro and Tailwind.
nemo
:
NEMO is a mobile data collection and analysis web application.
Nemoh
:
This project has migrated to GitLab: https://gitlab.com/lheea/Nemoh
// repository documentation
Was this content helpful?
★ 0
(0 ratings)
Select Rating:
★
★
★
★
★
Submit Feedback
Recent Feedback
×
Download README
Do you want to download the
README.md
file for
nemesis
?
Download (.md)
# Reward-Model Reward Model training framework for LLM RLHF. For in-depth understanding of Reward modeling, checkout our [blog](https://explodinggradients.com/) The word nemesis originally meant the distributor of fortune, neither good nor bad, simply in due proportion to each according to what was deserved. ### Quick Start * Inference ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer MODEL = "shahules786/Reward-model-gptneox-410M" model = AutoModelForSequenceClassification.from_pretrained(MODEL) tokenizer = AutoTokenizer.from_pretrained(MODEL) ``` * Training ```bash python src/training.py --config-name <your-config-name> ``` ## Contributions * All contributions are welcome. Checkout #issues
