McGill-NLP / length-generalizationLinks
Code for the paper "The Impact of Positional Encoding on Length Generalization in Transformers", NeurIPS 2023
☆136Updated last year
Alternatives and similar repositories for length-generalization
Users that are interested in length-generalization are comparing it to the libraries listed below
Sorting:
- Language models scale reliably with over-training and on downstream tasks☆97Updated last year
- [NeurIPS 2023] Learning Transformer Programs☆161Updated last year
- ☆74Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆81Updated 3 years ago
- Code repository for the c-BTM paper☆106Updated last year
- Efficient Transformers with Dynamic Token Pooling☆62Updated 2 years ago
- ☆53Updated last year
- Randomized Positional Encodings Boost Length Generalization of Transformers☆81Updated last year
- ☆65Updated 10 months ago
- ☆159Updated 2 years ago
- ☆83Updated last year
- Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Fl…☆75Updated 10 months ago
- Code for Zero-Shot Tokenizer Transfer☆133Updated 5 months ago
- Repository for the code of the "PPL-MCTS: Constrained Textual Generation Through Discriminator-Guided Decoding" paper, NAACL'22☆66Updated 2 years ago
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆68Updated 11 months ago
- Evaluation pipeline for the BabyLM Challenge 2023.☆76Updated last year
- ☆38Updated last year
- ☆45Updated last year
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆116Updated 2 years ago
- This is the implementation of the paper AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning (https://arxiv.org/abs/2205.1…☆132Updated last year
- RL algorithm: Advantage induced policy alignment☆65Updated last year
- ☆55Updated last year
- ☆86Updated last year
- ☆51Updated last year
- Multipack distributed sampler for fast padding-free training of LLMs☆194Updated 11 months ago
- Official Repository of Pretraining Without Attention (BiGS), BiGS is the first model to achieve BERT-level transfer learning on the GLUE …☆113Updated last year
- Recurrent Memory Transformer☆150Updated last year
- The Official Repository for "Bring Your Own Data! Self-Supervised Evaluation for Large Language Models"☆108Updated last year
- Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…☆54Updated last year
- ☆72Updated 2 years ago