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

    Neuronpedia

    INDEX
    Explanations

    adjective + noun combinations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     diriger
    -1.56
     exceedingly
    -1.54
     både
    -1.52
    Anybody
    -1.46
     solamente
    -1.45
     faciliter
    -1.43
     siè
    -1.35
     frequentemente
    -1.34
     genoux
    -1.32
    BOTH
    -1.31
    POSITIVE LOGITS
     in
    1.88
     with
    1.84
     at
    1.78
     that
    1.74
     and
    1.59
     during
    1.59
     difficult
    1.58
     relatively
    1.49
     this
    1.48
     really
    1.46
    Act Density 0.056%

    No Known Activations