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

    Neuronpedia

    INDEX
    Explanations

    defining or listing using punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     cheated
    0.51
     towed
    0.46
     wandered
    0.44
     blacksmith
    0.44
     diaries
    0.43
     burdens
    0.43
     concierge
    0.42
    ='/
    0.42
     continuo
    0.42
     duis
    0.41
    POSITIVE LOGITS
    Theorem
    0.48
     Assertion
    0.47
     Propositions
    0.44
    У
    0.44
    assertion
    0.43
     Theorem
    0.42
     Proposition
    0.41
    原
    0.41
    S
    0.41
    Proposition
    0.40
    Act Density 0.000%

    No Known Activations