CarperAI / decontamination
This repository contains code for cleaning your training data of benchmark data to help combat data snooping.
☆25Updated 2 years ago
Alternatives and similar repositories for decontamination:
Users that are interested in decontamination are comparing it to the libraries listed below
- A library for squeakily cleaning and filtering language datasets.☆47Updated last year
- QAmeleon introduces synthetic multilingual QA data using PaLM, a 540B large language model. This dataset was generated by prompt tuning P…☆34Updated last year
- ☆22Updated last year
- Repository containing the SPIN experiments on the DIBT 10k ranked prompts☆24Updated last year
- ☆20Updated last year
- Experiments with generating opensource language model assistants☆97Updated last year
- [WIP] Transformer to embed Danbooru labelsets☆13Updated last year
- Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit☆63Updated last year
- ☆44Updated 5 months ago
- Code for the examples presented in the talk "Training a Llama in your backyard: fine-tuning very large models on consumer hardware" given…☆14Updated last year
- a pipeline for using api calls to agnostically convert unstructured data into structured training data☆30Updated 7 months ago
- This is a new metric that can be used to evaluate faithfulness of text generated by LLMs. The work behind this repository can be found he…☆31Updated last year
- ☆24Updated last year
- ☆33Updated 10 months ago
- Plug-and-play Search Interfaces with Pyserini and Hugging Face☆31Updated last year
- QLoRA with Enhanced Multi GPU Support☆37Updated last year
- ☆36Updated this week
- ☆49Updated last year
- Utilities for Training Very Large Models☆58Updated 7 months ago
- ☆39Updated 2 years ago
- ☆48Updated last year
- 🤝 Trade any tensors over the network☆30Updated last year
- ☆37Updated last year
- Code repository for the c-BTM paper☆106Updated last year
- ☆38Updated last year
- ☆61Updated last year
- ☆43Updated 2 months ago
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated 2 years ago
- Experiments for efforts to train a new and improved t5☆77Updated last year
- Embedding Recycling for Language models☆38Updated last year