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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130421
    Prev
    Next
    INDEX
    Explanations

    The neuron fires whenever it encounters the word “safe” (whether in plain text, in quotes, or as part of an identifier).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    safe zone or haven

    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
    </h2>
    -2.34
     และ
    -2.34
     We
    -2.25
    鹮
    -2.22
    </strong>
    -2.16
    毉
    -2.16
    我能
    -2.13
    锇
    -2.11
    
    -2.11
     viviendo
    -2.08
    POSITIVE LOGITS
    ﮨ
    2.55
    啭
    2.31
    ꜆
    2.30
    箝
    2.27
    
    2.25
    跻
    2.25
     کنید
    2.25
     haven
    2.17
     یا
    2.14
    我们的
    2.14
    Activations Density 0.024%

    No Known Activations