graphcore-research / pytorch-tensor-trackerLinks
Flexibly track outputs and grad-outputs of torch.nn.Module.
☆13Updated last year
Alternatives and similar repositories for pytorch-tensor-tracker
Users that are interested in pytorch-tensor-tracker are comparing it to the libraries listed below
Sorting:
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆77Updated 10 months ago
- These papers will provide unique insightful concepts that will broaden your perspective on neural networks and deep learning☆48Updated last year
- Unofficial but Efficient Implementation of "Mamba: Linear-Time Sequence Modeling with Selective State Spaces" in JAX☆81Updated last year
- ☆37Updated last year
- JAX implementation of the Mistral 7b v0.2 model☆34Updated 11 months ago
- LoRA for arbitrary JAX models and functions☆136Updated last year
- Einsum-like high-level array sharding API for JAX☆34Updated 10 months ago
- JAX bindings for Flash Attention v2☆88Updated 10 months ago
- WIP☆93Updated 9 months ago
- Minimal but scalable implementation of large language models in JAX☆34Updated 7 months ago
- ☆78Updated 11 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆127Updated last year
- Language models scale reliably with over-training and on downstream tasks☆97Updated last year
- ☆33Updated 8 months ago
- ☆53Updated last year
- ☆29Updated 6 months ago
- A simple library for scaling up JAX programs☆137Updated 7 months ago
- Stick-breaking attention☆56Updated 2 months ago
- Experiment of using Tangent to autodiff triton☆79Updated last year
- supporting pytorch FSDP for optimizers☆79Updated 5 months ago
- Implementation of PSGD optimizer in JAX☆33Updated 5 months ago
- An implementation of the Llama architecture, to instruct and delight☆21Updated this week
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"☆73Updated 7 months ago
- ☆46Updated last year
- CUDA implementation of autoregressive linear attention, with all the latest research findings☆44Updated 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…☆54Updated last year
- If it quacks like a tensor...☆58Updated 6 months ago
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆44Updated 10 months ago
- ☆53Updated 8 months ago
- ☆29Updated 2 months ago