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

    Neuronpedia

    INDEX
    Explanations

    highly repeated symbols or special characters, indicating significant emphasis or markers in the text

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

    mathematical references

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    исленность
    -0.48
     مرئيه
    -0.46
    enumi
    -0.44
     Mend
    -0.42
     meny
    -0.40
     appui
    -0.39
    пенно
    -0.37
     enx
    -0.36
    พาะ
    -0.36
    __*/
    -0.36
    POSITIVE LOGITS
    parsedMessage
    0.63
     GenerationType
    0.56
    cerpt
    0.52
     fufficient
    0.52
     './../
    0.51
    oustic
    0.49
     Efq
    0.49
     Eſ
    0.48
    Massimo
    0.48
     gij
    0.47
    Act Density 0.059%

    No Known Activations