EleutherAI / nanoGPT-mupLinks
The simplest, fastest repository for training/finetuning medium-sized GPTs.
β166Updated 3 months ago
Alternatives and similar repositories for nanoGPT-mup
Users that are interested in nanoGPT-mup are comparing it to the libraries listed below
Sorting:
- A MAD laboratory to improve AI architecture designs π§ͺβ131Updated 10 months ago
- Understand and test language model architectures on synthetic tasks.β233Updated 3 weeks ago
- nanoGPT-like codebase for LLM trainingβ107Updated 5 months ago
- β91Updated last year
- supporting pytorch FSDP for optimizersβ83Updated 10 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT trainingβ132Updated last year
- β53Updated last year
- Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"β241Updated 4 months ago
- Normalized Transformer (nGPT)β192Updated 11 months ago
- Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"β84Updated 11 months ago
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)β192Updated last year
- πSmall Batch Size Training for Language Modelsβ63Updated 2 weeks ago
- β142Updated last month
- A set of Python scripts that makes your experience on TPU betterβ54Updated last month
- Language models scale reliably with over-training and on downstream tasksβ100Updated last year
- some common Huggingface transformers in maximal update parametrization (Β΅P)β85Updated 3 years ago
- Scaling is a distributed training library and installable dependency designed to scale up neural networks, with a dedicated module for trβ¦β64Updated 3 weeks ago
- β283Updated last year
- Some preliminary explorations of Mamba's context scaling.β216Updated last year
- Implementation of π₯₯ Coconut, Chain of Continuous Thought, in Pytorchβ179Updated 4 months ago
- β58Updated last year
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clustersβ130Updated 10 months ago
- β86Updated last year
- A fusion of a linear layer and a cross entropy loss, written for pytorch in triton.β69Updated last year
- β83Updated last year
- β102Updated 3 months ago
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmindβ129Updated last year
- β36Updated 8 months ago
- PyTorch-native post-training at scaleβ83Updated this week
- Mixture of A Million Expertsβ48Updated last year