© 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. 34-GEMMASCOPE-RES-131K
    7. 75791
    Prev
    Next
    INDEX
    Explanations

    protected areas and protection

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly responds to mentions of “Park” (and similar protected‐area names), effectively detecting references to parks or nature reserves.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     our
    -1.06
     their
    -1.00
     Gemeinschaft
    -0.94
    ajuku
    -0.91
     each
    -0.91
     antwoord
    -0.88
    口座
    -0.88
     the
    -0.86
     Weiterbildung
    -0.86
     using
    -0.85
    POSITIVE LOGITS
     Protects
    1.41
    Protecting
    1.36
     protected
    1.36
     protects
    1.36
    Protect
    1.33
     protecting
    1.27
    protect
    1.26
     protège
    1.25
    protection
    1.23
    Protection
    1.23
    Activations Density 0.060%

    No Known Activations