EleutherAI / nanoGPT-mup
The simplest, fastest repository for training/finetuning medium-sized GPTs.
โ83Updated last week
Related projects โ
Alternatives and complementary repositories for nanoGPT-mup
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT trainingโ112Updated 6 months ago
- A MAD laboratory to improve AI architecture designs ๐งชโ95Updated 6 months ago
- โ72Updated 4 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clustersโ104Updated last month
- Understand and test language model architectures on synthetic tasks.โ161Updated 6 months ago
- โ61Updated 2 months ago
- Scaling is a distributed training library and installable dependency designed to scale up neural networks, with a dedicated module for trโฆโ46Updated last week
- Experiments for efforts to train a new and improved t5โ76Updated 6 months ago
- code for training & evaluating Contextual Document Embedding modelsโ93Updated this week
- Collection of autoregressive model implementationโ66Updated last week
- Muon optimizer for neural networks: >30% extra sample efficiency, <3% wallclock overheadโ69Updated this week
- โ100Updated 3 months ago
- Language models scale reliably with over-training and on downstream tasksโ94Updated 7 months ago
- โ53Updated 9 months ago
- $100K or 100 Days: Trade-offs when Pre-Training with Academic Resourcesโ92Updated last week
- โ50Updated 5 months ago
- โ50Updated last week
- โ46Updated last month
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmindโ111Updated 2 months ago
- WIPโ89Updated 2 months ago
- Scalable neural net training via automatic normalization in the modular norm.โ119Updated 2 months ago
- Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (NeurIPS 2024)โ176Updated 5 months ago
- Implementation of the Llama architecture with RLHF + Q-learningโ156Updated 10 months ago
- โ105Updated this week
- โ121Updated this week
- โ50Updated last month
- โ39Updated 9 months ago
- some common Huggingface transformers in maximal update parametrization (ยตP)โ76Updated 2 years ago
- A set of Python scripts that makes your experience on TPU betterโ40Updated 4 months ago
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"โ84Updated 2 months ago