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

    Neuronpedia

    INDEX
    Explanations

    Greek words indicating actions or descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vutta
    0.25
     tathapi
    0.25
     Når
    0.25
     và
    0.25
     vadati
    0.24
     tasmim
    0.24
     Қ
    0.24
     vasena
    0.24
     Dacă
    0.24
     natthi
    0.23
    POSITIVE LOGITS
     α
    0.28
     σ
    0.24
     β
    0.23
     ε
    0.23
     θ
    0.23
     εξ
    0.23
     sp
    0.22
     δ
    0.22
     ω
    0.21
     δια
    0.21
    Act Density 0.001%

    No Known Activations