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

    Neuronpedia

    INDEX
    Explanations

    JSON object structure and associated property/value pairs

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

    Code snippets or serialized data

    np_acts-logits-general · gemini-2.0-flash

    JSON or code block delimiters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <eos>
    -0.74
    ful
    -0.50
    ↵↵↵
    -0.48
    ↵↵↵↵
    -0.47
    full
    -0.46
    and
    -0.43
    .
    -0.42
    ^
    -0.42
    <strong>
    -0.42
    nt
    -0.41
    POSITIVE LOGITS
     resourceCulture
    1.38
    InjectAttribute
    1.24
    __':
    1.22
    __":
    1.20
    GEBURTSDATUM
    1.19
    EndGlobalSection
    1.12
    ագրություններ
    1.07
    Erreferentziak
    1.06
     الحره
    1.06
     Taktlose
    1.03
    Act Density 0.020%

    No Known Activations