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

    Neuronpedia

    INDEX
    Explanations

    craft, ceph, 😴, GPT

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     conclude
    0.48
     accommodates
    0.46
    Ⅲ
    0.43
     concludes
    0.43
     accomplishes
    0.43
     conclure
    0.40
     गाँव
    0.39
     concede
    0.37
     Paintings
    0.37
    닝
    0.37
    POSITIVE LOGITS
     апре
    0.44
     maraming
    0.42
     inder
    0.40
     уя
    0.39
     vilket
    0.38
     ander
    0.38
     ane
    0.38
    €”
    0.38
     כן
    0.37
    elto
    0.37
    Act Density 0.001%

    No Known Activations