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

    Neuronpedia

    INDEX
    Explanations

    version control and tracking changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ="..\
    0.47
     PATCH
    0.41
     committed
    0.40
     COMMIT
    0.40
    ಷ್ಟು
    0.39
     interactive
    0.39
    commit
    0.39
     commits
    0.39
     commit
    0.38
     execute
    0.38
    POSITIVE LOGITS
    Versions
    0.60
     revisions
    0.58
     versions
    0.56
    Revision
    0.55
     Versions
    0.55
     versões
    0.54
    versions
    0.53
    version
    0.52
     versionName
    0.52
     버전
    0.51
    Act Density 0.009%

    No Known Activations