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

    Neuronpedia

    INDEX
    Explanations

    phrases related to competition or comparison

    oai_token-act-pair · gpt-3.5-turbo

    complex sentence structures involving a range of clauses or phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    GF
    -0.80
    UF
    -0.72
    ounded
    -0.68
    ega
    -0.67
    OME
    -0.64
    USS
    -0.64
    IVER
    -0.63
    ForgeModLoader
    -0.63
    ound
    -0.62
    Sov
    -0.62
    POSITIVE LOGITS
     lest
    1.04
     namely
    0.90
     preferably
    0.89
     albeit
    0.87
     but
    0.82
     whereas
    0.80
     depending
    0.75
     although
    0.73
     respectively
    0.73
     ie
    0.73
    Act Density 0.498%

    No Known Activations