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

    Neuronpedia

    INDEX
    Explanations

    phrases relating to grounding and connection to reality

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

    grounded, anchored, rooted

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    BufferException
    -0.71
     Administrativna
    -0.70
    ValueStyle
    -0.69
    évaluateur
    -0.69
    Tembelea
    -0.66
    niſſe
    -0.66
    verwijspagina
    -0.66
     Савезне
    -0.66
    SharedDtor
    -0.65
    ſchaft
    -0.65
    POSITIVE LOGITS
     rooted
    0.88
     grounded
    0.84
     grounding
    0.82
     anchored
    0.74
    rooted
    0.70
     anchoring
    0.68
    Grounded
    0.66
     anchor
    0.60
     anchors
    0.58
     Ground
    0.56
    Act Density 0.041%

    No Known Activations