Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    python code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     starving
    0.86
     Kodak
    0.86
     Sov
    0.85
     Emperor
    0.84
     Fed
    0.80
     nucleation
    0.79
     handing
    0.78
     Woodrow
    0.76
     Dick
    0.75
     eV
    0.75
    POSITIVE LOGITS
    python
    2.20
    java
    1.62
    Python
    1.57
     python
    1.55
    javascript
    1.54
    cpp
    1.50
    bash
    1.49
    json
    1.41
    html
    1.41
    php
    1.40
    Act Density 0.070%

    No Known Activations