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

    Neuronpedia

    INDEX
    Explanations

    prepositions and directions

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interessante
    0.54
    Healthcare
    0.52
     fart
    0.52
    惬
    0.50
     consistente
    0.49
     trustworthiness
    0.48
     interesante
    0.48
    PROP
    0.48
     interesting
    0.47
     endExpNow
    0.47
    POSITIVE LOGITS
     Into
    0.53
     Own
    0.51
     кван
    0.48
     Toward
    0.47
     into
    0.46
     Пол
    0.45
    ových
    0.44
     From
    0.44
     vào
    0.43
     With
    0.43
    Act Density 0.000%

    No Known Activations