ml-jku / SDLGLinks
SDLG is an efficient method to accurately estimate aleatoric semantic uncertainty in LLMs
☆27Updated last year
Alternatives and similar repositories for SDLG
Users that are interested in SDLG are comparing it to the libraries listed below
Sorting:
- ☆69Updated last year
- ☆82Updated last year
- DoG is SGD's Best Friend: A Parameter-Free Dynamic Step Size Schedule☆63Updated 2 years ago
- ☆131Updated last month
- Official implementation of "GPT or BERT: why not both?"☆57Updated last month
- ☆53Updated 2 years ago
- Modalities, a PyTorch-native framework for distributed and reproducible foundation model training.☆84Updated this week
- Minimum Bayes Risk Decoding for Hugging Face Transformers☆58Updated last year
- Official implementation of "BERTs are Generative In-Context Learners"☆32Updated 5 months ago
- Simple and scalable tools for data-driven pretraining data selection.☆25Updated 2 months ago
- Sparse and discrete interpretability tool for neural networks☆63Updated last year
- PyTorch library for Active Fine-Tuning☆89Updated 6 months ago
- Implementation of the BatchTopK activation function for training sparse autoencoders (SAEs)☆45Updated last month
- Interpretating the latent space representations of attention head outputs for LLMs☆34Updated last year
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆82Updated 10 months ago
- A repository containing the code for translating popular LLM benchmarks to German.☆28Updated 2 years ago
- nanoGPT-like codebase for LLM training☆102Updated 3 months ago
- ☆81Updated 6 months ago
- Efficient Transformers with Dynamic Token Pooling☆63Updated 2 years ago
- ☆56Updated 10 months ago
- ☆36Updated 2 years ago
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆53Updated last year
- ☆104Updated 6 months ago
- State-of-the-art paired encoder and decoder models (17M-1B params)☆43Updated 3 weeks ago
- Simple-to-use scoring function for arbitrarily tokenized texts.☆45Updated 6 months ago
- ☆27Updated 2 years ago
- A Toolkit for Distributional Control of Generative Models☆73Updated 3 weeks ago
- Erasing concepts from neural representations with provable guarantees☆232Updated 7 months ago
- ☆166Updated 2 years ago
- Official Repository of Pretraining Without Attention (BiGS), BiGS is the first model to achieve BERT-level transfer learning on the GLUE …☆114Updated last year