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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating feelings of something being wrong or problematic

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

    Something is wrong or has gone wrong

    np_acts-logits-general · gemini-2.0-flash

    something wrong

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    leid
    -0.48
     inol
    -0.47
     Swal
    -0.46
     completos
    -0.46
    ailles
    -0.42
     completo
    -0.42
    ourge
    -0.41
    ToUse
    -0.41
    Complete
    -0.40
    RunWith
    -0.40
    POSITIVE LOGITS
    ]")]
    1.16
     happening
    1.10
     amiss
    1.10
     wrong
    1.00
     happened
    0.92
     مشين
    0.91
     happen
    0.90
     transpired
    0.88
    HomeAsUpEnabled
    0.88
    happen
    0.84
    Act Density 0.224%

    No Known Activations