✳flâneur — a map of the web's best reading
Add NTK-Aware interpolation "by parts" correction by bloc97 · Pull Request #1 · jquesnelle/scaled-rope
github.com · 825 words · saved by 1 readers
This PR adds the new and improved "by parts" correction to the NTK-aware interpolation method. This corrected method improves from previous methods fourfold: Decreases PPL in all context lengths w...
Explore this link on the map →saved by
related reading
- Optimizing Cubature for Efficient Integration of Subspace Deformationscs.cornell.edu
- Operator Aproach to Tangent Vector Processing.pdfgeometrica.saclay.inria.fr
- Jacobian Lens – Qwen3.6-27B ｜ Neuronpedianeuronpedia.org
- GitHub - NVlabs/AutoGaze: AutoGaze automatically removes redundant patches in a video, reducing #tokens in ViT/MLLM by 4x-100x. · GitHubgithub.com
- Regularized by Score Matching (LeCun)papers.nips.cc
- python - Which kind of interpolation best for resizing image? - Stack Overflowstackoverflow.com
- GitHub - inverse-scaling/prize: A prize for finding tasks that cause large language models to show inverse scaling · GitHubgithub.com
- Innovation Prior - reconstructed reasoning tracesbohanlyu.com
- GitHub - Masao-Taketani/FOTS_OCR: TensorFlow Implementation of FOTS, Fast Oriented Text Spotting with a Unified Network. · GitHubgithub.com
- slips/slip-0039.md at master · satoshilabs/slips · GitHubgithub.com
- Plus 2: Luyện đọc điền và đọc hiểu chuyên sâu (2024)ngoaingu24h.vn
- Koinakoina.wilhelmlab.org