© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70858
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by conditionals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ünün
    -0.97
     trámites
    -0.96
    étaires
    -0.96
     CHOOSE
    -0.92
    probably
    -0.91
    しかも
    -0.91
     aina
    -0.91
     gagner
    -0.89
     PRIM
    -0.89
     cardin
    -0.88
    POSITIVE LOGITS
     If
    1.77
     if
    1.76
     otherwise
    1.41
     Otherwise
    1.20
    If
    1.16
     or
    1.08
     then
    1.05
     Sadly
    1.05
     which
    1.02
     because
    1.02
    Activations Density 0.030%

    No Known Activations