Language Breakdown
Lines of code distribution across 10 owned repositories
I-Shaped Developer
I-shapedSpecialist — deep expertise in Python
Collaboration Network
Global Impact visualization
Repos
12
PRs
0
Growth
+18%
Top Collaborators
No collaborator data yet.
Coding Streak
Contribution activity over the past year
Motion-TimeSpace
@Martin123132
挥手的骑士
@yihuishou
MindIntels
@MindIntels
schmug
@schmug
niloc
@colindaymond
Top Repositories
Sticky-block topology lottery scheduler for transformer fine-tuning.
Adaptive Event-Control Scheduler (AECS) — a state-aware, event-driven LR scheduler for PyTorch
Real WiFi in your VM. No dongle required.
Can a static SAE feature atlas of a frozen LLM be driven as a runnable, legible surrogate? Open research.
Mechanistic Interpretability toolkit for Vision-Language-Action models
GWIQ-Atlas: is a brain-atlasing and model-interpretability suite that combines per-layer census, compliance behaviour tracing, SAE features, and quantization analyses for LLMs.
Self-tuning SAE trainer: train a sparse autoencoder on every layer of any model in one unattended run. No per-layer retuning. Using an augmented-Lagrangian controller hits the sparsity target mid-run instead of restarting on failure.
An independent, self-contained library implementing 10 state-of-the-art LLM post-training quantization (PTQ) and fine-tuning algorithms.
A public operational registry of ML tools, methods, papers, models, and training techniques — structured for fine-tuning advice retrieval. Zero PII. Source-backed only.
Open Source Impact
Contributions to external projects
