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

    Neuronpedia

    INDEX
    Explanations

    instances of the prefixes and suffixes "se-", "in-", and "-se."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    sel
    -0.20
    swers
    -0.18
    rag
    -0.17
    tar
    -0.15
    sWith
    -0.15
    shaft
    -0.15
    errer
    -0.15
    dy
    -0.15
    缸æīĭ
    -0.15
    thag
    -0.15
    POSITIVE LOGITS
    parable
    0.24
    attle
    0.24
    curities
    0.23
    ismic
    0.22
    bastian
    0.22
    parate
    0.21
    gregation
    0.20
    paration
    0.20
    ATTLE
    0.20
    parated
    0.19
    Act Density 0.052%

    No Known Activations