erogol / BlaGPTLinks
Experimental playground for benchmarking language model (LM) architectures, layers, and tricks on smaller datasets. Designed for flexible experimentation and exploration.
☆92Updated 3 weeks ago
Alternatives and similar repositories for BlaGPT
Users that are interested in BlaGPT are comparing it to the libraries listed below
Sorting:
- Repository for "TESS-2: A Large-Scale, Generalist Diffusion Language Model"☆53Updated 10 months ago
- Block Transformer: Global-to-Local Language Modeling for Fast Inference (NeurIPS 2024)☆163Updated 8 months ago
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆56Updated 10 months ago
- A byte-level decoder architecture that matches the performance of tokenized Transformers.☆66Updated last year
- Griffin MQA + Hawk Linear RNN Hybrid☆89Updated last year
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆132Updated 2 months ago
- DPO, but faster 🚀☆46Updated last year
- Official repository for the paper "SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention"☆102Updated last year
- Implementation of a Light Recurrent Unit in Pytorch☆49Updated last year
- Implementation of the proposed Adam-atan2 from Google Deepmind in Pytorch☆134Updated 2 months ago
- Tiled Flash Linear Attention library for fast and efficient mLSTM Kernels.☆81Updated last month
- Implementation of Infini-Transformer in Pytorch☆112Updated last year
- Randomized Positional Encodings Boost Length Generalization of Transformers☆82Updated 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…☆53Updated 2 years ago
- MiSS is a novel PEFT method that features a low-rank structure but introduces a new update mechanism distinct from LoRA, achieving an exc…☆25Updated 2 months ago
- Official code release for "SuperBPE: Space Travel for Language Models"☆77Updated last month
- Attempt to make multiple residual streams from Bytedance's Hyper-Connections paper accessible to the public☆124Updated this week
- ☆91Updated last year
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.☆74Updated last year
- Here we will test various linear attention designs.☆62Updated last year
- RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best…☆59Updated 9 months ago
- Collection of autoregressive model implementation☆85Updated this week
- research impl of Native Sparse Attention (2502.11089)☆63Updated 10 months ago
- Explorations into adversarial losses on top of autoregressive loss for language modeling☆41Updated 2 weeks ago
- Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch☆181Updated 6 months ago
- An unofficial pytorch implementation of 'Efficient Infinite Context Transformers with Infini-attention'☆54Updated last year
- Some personal experiments around routing tokens to different autoregressive attention, akin to mixture-of-experts☆121Updated last year
- My attempts at applying Soundstream design on learned tokenization of text and then applying hierarchical attention to text generation☆89Updated last year
- ☆54Updated last year
- Explorations into the recently proposed Taylor Series Linear Attention☆100Updated last year