Authentrics.ai

company

Verified

https://authentrics.ai

AI & ML interests

We build tools to understand how models change during training, identify where regressions and unwanted behaviors emerge, localize meaningful changes within the model, and correct or remove learned behavior without full retraining.Our work spans training dynamics, model interpretability, machine unlearning, training-free model optimization, and AI governance.

Recent Activity

seanhagstrom updated a model 14 days ago

authentrics/ztom-gemma3-1b-recovery

seanhagstrom updated a model 14 days ago

authentrics/pii-removal-medical-llama32

seanhagstrom updated a Space 14 days ago

authentrics/README

View all activity

Articles

Accelerating model refinement: automating Fine-Tuning through checkpoint interpolation with Authentrics Zero Training Optimization & Maintenance (Z-TOM)

updated 2 models 14 days ago

authentrics/ztom-gemma3-1b-recovery

Text Generation • Updated 14 days ago • 1

authentrics/pii-removal-medical-llama32

Text Generation • Updated 14 days ago

updated a Space 14 days ago

README

We are Authentrics.ai

updated 4 models 14 days ago

authentrics/nemotron-training-dynamics

Text Generation • Updated 14 days ago

authentrics/mnist-badnets-repaired

Image Classification • Updated 14 days ago

authentrics/mnist-badnets-poisoned

Image Classification • Updated 14 days ago • 1

authentrics/mixtral-expert-routing

Text Generation • Updated 14 days ago

brandon-authentrics

published an article 22 days ago

Article

Accelerating model refinement: automating Fine-Tuning through checkpoint interpolation with Authentrics Zero Training Optimization & Maintenance (Z-TOM)

authentrics

•

22 days ago

updated a Space 23 days ago

README

We are Authentrics.ai

updated 5 collections 23 days ago

Recover a Capability Lost During Fine-Tuning

Sequential fine-tuning quietly degraded a capability the model already had. Authentrics restores it — no retraining from scratch. • 1 item • Updated 23 days ago

Removing Memorized PII From a Fine-Tuned Model

A model fine-tuned on data containing PII memorized it. Authentrics locates and removes that influence — no retraining from scratch. • 1 item • Updated 23 days ago

Reading How a Mixture-of-Experts Model Routes Its Inputs

Reading how Mixtral-8x7B routes activations, and how that structure emerges across training — without moving weights off your hardware. • 1 item • Updated 23 days ago

Diagnosing What Changed Across a Model's Training Run

Tracking parameter and activation drift across the NVIDIA Nemotron Cascade 8B checkpoint lineage to see what changed, where, and when. • 1 item • Updated 23 days ago

Finding and Removing a Neural Network Backdoor

A BadNets backdoor is planted in one MNIST training epoch, localized via checkpoint drift, and removed with Authentrics — no retraining from scratch. • 2 items • Updated 23 days ago

updated a model 23 days ago

authentrics/ztom-gemma3-1b-recovery

Text Generation • Updated 14 days ago • 1

published a model 23 days ago

authentrics/ztom-gemma3-1b-recovery

Text Generation • Updated 14 days ago • 1

updated a model 23 days ago

authentrics/pii-removal-medical-llama32

Text Generation • Updated 14 days ago

published a model 23 days ago

authentrics/pii-removal-medical-llama32

Text Generation • Updated 14 days ago

updated a model 23 days ago

authentrics/nemotron-training-dynamics

Text Generation • Updated 14 days ago