LucasPrietoAl / grokking-at-the-edge-of-numerical-stabilityLinks
☆98Updated 5 months ago
Alternatives and similar repositories for grokking-at-the-edge-of-numerical-stability
Users that are interested in grokking-at-the-edge-of-numerical-stability are comparing it to the libraries listed below
Sorting:
- supporting pytorch FSDP for optimizers☆82Updated 6 months ago
- ☆53Updated last year
- The simplest, fastest repository for training/finetuning medium-sized GPTs.☆134Updated this week
- Normalized Transformer (nGPT)☆183Updated 7 months ago
- ☆81Updated last year
- NanoGPT-speedrunning for the poor T4 enjoyers☆66Updated 2 months ago
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"☆101Updated 6 months ago
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆101Updated 3 months ago
- ☆78Updated 11 months ago
- DeMo: Decoupled Momentum Optimization☆188Updated 6 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆126Updated 6 months ago
- Implementations of attention with the softpick function, naive and FlashAttention-2☆77Updated last month
- Token Omission Via Attention☆127Updated 8 months ago
- Understand and test language model architectures on synthetic tasks.☆217Updated 2 weeks ago
- Mixture of A Million Experts☆46Updated 10 months ago
- EvaByte: Efficient Byte-level Language Models at Scale☆101Updated 2 months ago
- A MAD laboratory to improve AI architecture designs 🧪☆120Updated 6 months ago
- Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun☆52Updated 3 months ago
- ☆61Updated 7 months ago
- Collection of autoregressive model implementation☆85Updated last month
- WIP☆93Updated 10 months ago
- Fast, Modern, Memory Efficient, and Low Precision PyTorch Optimizers☆94Updated this week
- research impl of Native Sparse Attention (2502.11089)☆54Updated 4 months ago
- Efficient optimizers☆220Updated this week
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆127Updated last year
- RWKV-7: Surpassing GPT☆91Updated 7 months ago
- ☆49Updated last year
- ☆56Updated 3 months ago
- ☆188Updated 6 months ago
- Focused on fast experimentation and simplicity☆74Updated 5 months ago