TheMody / No-learning-rates-needed-Introducing-SALSA-Stable-Armijo-Line-Search-AdaptationLinks
SaLSa Optimizer implementation (No learning rates needed)
☆31Updated last month
Alternatives and similar repositories for No-learning-rates-needed-Introducing-SALSA-Stable-Armijo-Line-Search-Adaptation
Users that are interested in No-learning-rates-needed-Introducing-SALSA-Stable-Armijo-Line-Search-Adaptation are comparing it to the libraries listed below
Sorting:
- Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"☆101Updated 6 months ago
- An implementation of PSGD Kron second-order optimizer for PyTorch☆92Updated 3 months ago
- PyTorch implementation of models from the Zamba2 series.☆184Updated 5 months ago
- ☆70Updated 10 months ago
- Efficiently discovering algorithms via LLMs with evolutionary search and reinforcement learning.☆103Updated this week
- ☆81Updated last year
- Official implementation of the paper "Linear Transformers with Learnable Kernel Functions are Better In-Context Models"☆161Updated 6 months ago
- ICLR 2025 - official implementation for "I-Con: A Unifying Framework for Representation Learning"☆102Updated 3 weeks ago
- The AdEMAMix Optimizer: Better, Faster, Older.☆183Updated 10 months ago
- Getting crystal-like representations with harmonic loss☆192Updated 3 months ago
- This repository contains a better implementation of Kolmogorov-Arnold networks☆62Updated last month
- A HuggingFace compatible Small Language Model trainer.☆75Updated 5 months ago
- Official implementation of the paper: "ZClip: Adaptive Spike Mitigation for LLM Pre-Training".☆130Updated 2 weeks ago
- σ-GPT: A New Approach to Autoregressive Models☆65Updated 11 months ago
- Official repository for the paper "NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networks". This rep…☆59Updated 8 months ago
- Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind☆127Updated 10 months ago
- $100K or 100 Days: Trade-offs when Pre-Training with Academic Resources☆140Updated last month
- ☆30Updated 9 months ago
- Implementations of attention with the softpick function, naive and FlashAttention-2☆80Updated 2 months ago
- A byte-level decoder architecture that matches the performance of tokenized Transformers.☆64Updated last year
- DeMo: Decoupled Momentum Optimization☆189Updated 7 months ago
- Official implementation of MetaTree: Learning a Decision Tree Algorithm with Transformers☆112Updated 10 months ago
- ☆134Updated 11 months ago
- ☆98Updated 5 months ago
- Tokun to can tokens☆18Updated last month
- FlashRNN - Fast RNN Kernels with I/O Awareness☆92Updated last month
- Implementation of Mind Evolution, Evolving Deeper LLM Thinking, from Deepmind☆55Updated last month
- Collection of autoregressive model implementation☆85Updated 2 months ago
- Gradient Boosting Reinforcement Learning (GBRL)☆116Updated last month
- Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients.☆198Updated last year