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

    Neuronpedia

    INDEX
    Explanations

    words reflecting deterioration or decline

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

    complete decay or failure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    InjectMocks
    -0.41
     <>",
    -0.37
    },[])
    -0.36
    CLE
    -0.34
    wüns
    -0.32
     CLE
    -0.31
    ()];
    -0.31
    BagConstraints
    -0.31
     monta
    -0.31
     Cl
    -0.31
    POSITIVE LOGITS
     miserably
    0.65
    tanleria
    0.64
     completely
    0.56
     nakalista
    0.52
     emotionally
    0.51
     horribly
    0.51
     hopelessly
    0.50
     Completely
    0.50
    sizeCache
    0.50
     complètement
    0.50
    Act Density 0.042%

    No Known Activations