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

    Neuronpedia

    INDEX
    Explanations

    expressions of strong beliefs and commitments

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

    positive descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵↵
    -0.34
    Hold
    -0.32
     as
    -0.31
     called
    -0.30
     话
    -0.30
    landı
    -0.30
    no
    -0.29
    verhält
    -0.29
     related
    -0.28
    related
    -0.28
    POSITIVE LOGITS
     Савезне
    0.89
     CreateTagHelper
    0.76
    <unused47>
    0.76
    <unused51>
    0.76
    <unused41>
    0.76
    [@BOS@]
    0.76
    <unused68>
    0.76
    <unused80>
    0.76
    <unused79>
    0.76
    <unused1>
    0.75
    Act Density 0.100%

    No Known Activations