kernelmachine / cbtm
Code repository for the c-BTM paper
☆106Updated last year
Alternatives and similar repositories for cbtm
Users that are interested in cbtm are comparing it to the libraries listed below
Sorting:
- Multipack distributed sampler for fast padding-free training of LLMs☆189Updated 9 months ago
- Experiments for efforts to train a new and improved t5☆77Updated last year
- Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit☆63Updated last year
- TART: A plug-and-play Transformer module for task-agnostic reasoning☆197Updated last year
- Language models scale reliably with over-training and on downstream tasks☆97Updated last year
- A repository for transformer critique learning and generation☆90Updated last year
- ☆38Updated last year
- Code for the paper "The Impact of Positional Encoding on Length Generalization in Transformers", NeurIPS 2023☆135Updated last year
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated 2 years ago
- The GitHub repo for Goal Driven Discovery of Distributional Differences via Language Descriptions☆69Updated 2 years ago
- Positional Skip-wise Training for Efficient Context Window Extension of LLMs to Extremely Length (ICLR 2024)☆205Updated 11 months ago
- ☆72Updated last year
- ☆33Updated 10 months ago
- [ICLR 2023] Guess the Instruction! Flipped Learning Makes Language Models Stronger Zero-Shot Learners☆115Updated 8 months ago
- ☆94Updated last year
- Experiments on speculative sampling with Llama models☆126Updated last year
- Spherical Merge Pytorch/HF format Language Models with minimal feature loss.☆121Updated last year
- ☆120Updated 7 months ago
- Multi-Domain Expert Learning☆67Updated last year
- Experiments with generating opensource language model assistants☆97Updated 2 years ago
- Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Fl…☆72Updated 8 months ago
- SILO Language Models code repository☆81Updated last year
- Code for the arXiv preprint "The Unreasonable Effectiveness of Easy Training Data"☆47Updated last year
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆220Updated last year
- some common Huggingface transformers in maximal update parametrization (µP)☆80Updated 3 years ago
- Replicating O1 inference-time scaling laws☆85Updated 5 months ago
- ☆125Updated last year
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models" [AISTATS …☆59Updated 7 months ago
- Evaluating LLMs with CommonGen-Lite☆90Updated last year
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆126Updated 5 months ago