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

    Neuronpedia

    INDEX
    Explanations

    You followed by role/capability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Patch
    0.71
    Patch
    0.68
     Indigo
    0.65
     patched
    0.64
    ाये
    0.62
     patching
    0.60
     IP
    0.59
     Seven
    0.59
     laugh
    0.59
     seventy
    0.59
    POSITIVE LOGITS
    baran
    0.69
    ungee
    0.68
    rit
    0.68
    san
    0.65
     Tn
    0.63
     میٹر
    0.61
    OnInit
    0.61
    rita
    0.60
    ρίς
    0.60
    âns
    0.60
    Act Density 0.175%

    No Known Activations