Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing

  • Log In
  • Sign Up

LoganResearch
/
cfhot-weights

Text Classification
PyTorch
Safetensors
English
behavioral-detection
hidden-state-probing
per-token-classification
cross-architecture
holonomy-transformer
control-field
AI-safety
probes
Model card Files Files and versions
xet
Community
cfhot-weights
478 MB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 18 commits
LoganResearch's picture
LoganResearch
Fix: correct enhancement probe results to match model card
e5e36d6 3 months ago
  • code
    🧠 Full weight release: 9 probes × 3 architectures + production adapter + training code 3 months ago
  • cognitive
    🧠 Full weight release: 9 probes × 3 architectures + production adapter + training code 3 months ago
  • production
    🧠 Full weight release: 9 probes × 3 architectures + production adapter + training code 3 months ago
  • results
    🧠 Full weight release: 9 probes × 3 architectures + production adapter + training code 3 months ago
  • suppression
    🧠 Full weight release: 9 probes × 3 architectures + production adapter + training code 3 months ago
  • .gitattributes
    1.58 kB
    add model card banner 3 months ago
  • README.md
    5.46 kB
    Fix: correct enhancement probe results to match model card 3 months ago
  • cfhot_model_card.png
    4.39 MB
    xet
    Upload cfhot_model_card.png 3 months ago
  • inference.py
    12.1 kB
    add universal inference loader — works with all probes 3 months ago
  • requirements.txt
    54 Bytes
    add requirements.txt 3 months ago
  • run.py
    17.9 kB
    Feature: Self-aware interactive chat - model senses its own steering 3 months ago