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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating understanding or clarification in conversation

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

    seeking understanding or clarification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     témoig
    -0.71
    ioutil
    -0.68
    \{\\
    -0.65
     BrowserModule
    -0.65
    脚注の使い方
    -0.65
    enumii
    -0.63
    ロウィン
    -0.63
    ðsíða
    -0.62
    gines
    -0.61
    ſſung
    -0.61
    POSITIVE LOGITS
    ?
    0.46
     noDo
    0.35
    Bref
    0.30
     Remember
    0.29
    tså
    0.28
    ?"
    0.28
    Remember
    0.28
    ?'
    0.28
     hence
    0.27
    ?”
    0.26
    Act Density 0.011%

    No Known Activations