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

    Neuronpedia

    INDEX
    Explanations

    edges of buildings and structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     langfrist
    0.67
     matemática
    0.64
     mentored
    0.59
     genomics
    0.58
     cognitiva
    0.57
     saudável
    0.57
     decisão
    0.55
     współpr
    0.55
     fluency
    0.54
     algorithmic
    0.54
    POSITIVE LOGITS
     wooden
    0.82
     walls
    0.82
     bungalows
    0.77
     railings
    0.77
     veranda
    0.76
     courtyard
    0.75
     fences
    0.75
     roofs
    0.75
     embank
    0.74
     fence
    0.73
    Act Density 0.070%

    No Known Activations