© 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. 117874
    Prev
    Next
    INDEX
    Explanations

    Event Description

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

    The neuron is keyed to the section‐heading and boilerplate tokens you see in privacy‐policy/legal documents (e.g. “Privacy Policy,” “What information do we collect?”, “You can verify…,” etc.).

    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
     five
    -0.98
     eight
    -0.90
     more
    -0.89
     less
    -0.87
     during
    -0.87
     not
    -0.87
     because
    -0.84
     toscana
    -0.82
     at
    -0.81
     starts
    -0.81
    POSITIVE LOGITS
     TIN
    1.05
     PURE
    0.99
     SPECIFIC
    0.99
     Tn
    0.95
    跟著
    0.93
     FTC
    0.93
     bụ
    0.92
     Uttarakhand
    0.91
     trebui
    0.91
    itum
    0.91
    Activations Density 0.021%

    No Known Activations