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

    Neuronpedia

    INDEX
    Explanations

    code-related terms associated with decoding operations

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

    decoding structured data fields

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     itemName
    -0.55
     parten
    -0.48
    Household
    -0.47
     ſta
    -0.46
    ніципа
    -0.46
    husband
    -0.46
     axon
    -0.45
    overhead
    -0.45
    Yan
    -0.44
     ſel
    -0.44
    POSITIVE LOGITS
    decode
    3.03
     decode
    2.13
    Decode
    1.89
     Decode
    1.81
     decoding
    1.46
     decoded
    1.41
    decoded
    1.36
    解码
    1.19
    Decoding
    1.15
    Decoder
    1.11
    Act Density 0.003%

    No Known Activations