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

    Neuronpedia

    INDEX
    Explanations

    conflicting or divergent paths

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     distinguishing
    0.40
     distinction
    0.39
    ເທ
    0.38
     Distinction
    0.37
     RO
    0.36
     நிற
    0.36
    य्
    0.35
     musculaire
    0.35
     Prote
    0.35
     Discussion
    0.35
    POSITIVE LOGITS
     incompatible
    1.07
     conflicting
    0.84
     incompat
    0.82
     incompatibility
    0.80
     orthogonal
    0.78
     divergent
    0.75
     disconnect
    0.73
     conflict
    0.73
    disconnect
    0.73
     clashes
    0.71
    Act Density 0.011%

    No Known Activations