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

    Neuronpedia

    INDEX
    Explanations

    applies to changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝗻
    0.38
     खोले
    0.37
     Pinn
    0.36
     nale
    0.35
     الفترة
    0.35
     pint
    0.35
    ipient
    0.35
    मोबाइल
    0.35
     sôi
    0.34
     हार्ट
    0.34
    POSITIVE LOGITS
     patch
    1.35
     diff
    1.30
    patch
    1.30
     patches
    1.23
    Patch
    1.20
     Patch
    1.16
    diff
    1.16
    Diff
    1.11
     patching
    1.08
     Diff
    1.05
    Act Density 0.171%

    No Known Activations