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

    Neuronpedia

    INDEX
    Explanations

    abstract representation, symbolism, metaphor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    opre
    0.49
    counts
    0.46
    study
    0.46
    negl
    0.46
    teach
    0.45
    uk
    0.45
    ovač
    0.45
    pro
    0.44
    ssel
    0.44
    assess
    0.44
    POSITIVE LOGITS
     metaphor
    0.60
     뭔가
    0.52
     metaphorical
    0.52
    或者是
    0.50
    或是
    0.50
    某种
    0.48
     allusion
    0.48
     Poetry
    0.47
    หรือ
    0.46
     symbolism
    0.46
    Act Density 0.280%

    No Known Activations