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

    Neuronpedia

    INDEX
    Explanations

    structured labels and descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     base
    0.60
      
    0.60
     around
    0.58
     either
    0.56
     (
    0.55
     included
    0.54
     linked
    0.54
     access
    0.53
     about
    0.53
     half
    0.53
    POSITIVE LOGITS
     prachtige
    0.77
    varez
    0.76
     vrouwen
    0.76
     docentes
    0.75
     menino
    0.75
    Pourquoi
    0.74
    <unused381>
    0.74
     sabiduría
    0.74
    thisStudent
    0.73
     televisie
    0.73
    Act Density 0.001%

    No Known Activations