amazon-science/ContraCLM

[ACL 2023] Code for ContraCLM: Contrastive Learning For Causal Language Model

29
/ 100
Experimental

This project helps machine learning engineers pre-train causal language models like GPT2 and CodeGen for better performance on various NLP and programming tasks. It takes in large natural language or programming language datasets and outputs a more robust, discriminative language model. Data scientists and ML researchers who work with large language models would use this to enhance their models beyond simple text generation.

No commits in the last 6 months.

Use this if you are a machine learning engineer or researcher looking to improve the representation quality and discriminative power of your decoder-only language models for tasks beyond just text generation.

Not ideal if you are looking for a pre-trained model to use directly in an application, as this is a framework for pre-training models.

natural-language-processing large-language-models deep-learning code-generation model-pretraining
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

35

Forks

2

Language

Python

License

Apache-2.0

Last pushed

Dec 20, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/amazon-science/ContraCLM"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.