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

    Neuronpedia

    INDEX
    Explanations

    items used in construction and household applications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Long, thin objects

    np_acts-logits-general · gemini-2.0-flash

    pointy objects, inserted and tipped

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    viewDid
    -0.68
     panels
    -0.56
     pouches
    -0.55
    decken
    -0.53
     Panels
    -0.52
     πλα
    -0.52
    Panels
    -0.50
    Layer
    -0.49
     blanket
    -0.49
    Panel
    -0.49
    POSITIVE LOGITS
     tip
    1.30
     Tip
    1.01
     tips
    0.99
    Tip
    0.96
     tipped
    0.96
    tip
    0.93
     punta
    0.89
    tipped
    0.88
     shaft
    0.87
     pointy
    0.86
    Act Density 0.610%

    No Known Activations