© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 70989
    Prev
    Next
    INDEX
    Explanations

    reporting by officials

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

    The neuron activates on words that attribute statements to an authority (e.g. “official,” “senior,” “government,” “intelligence”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     a
    -2.23
     ziemlich
    -1.55
     It
    -1.52
     What
    -1.52
     After
    -1.47
     There
    -1.41
     feiert
    -1.37
     beschreibt
    -1.36
     Vereine
    -1.36
     Surely
    -1.36
    POSITIVE LOGITS
     said
    1.91
     further
    1.69
    </i>
    1.67
     >/
    1.58
     omnes
    1.58
     cnn
    1.48
     omnium
    1.48
    eronau
    1.46
     }}(
    1.45
    ​​​
    1.41
    Activations Density 0.030%

    No Known Activations