© 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. 22-GEMMASCOPE-RES-131K
    7. 128860
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on terms relating to surveillance and documentation—words about monitoring, recording interactions, and gathering evidence.

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

    evidence and documentation

    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
    方には
    -1.09
    🏏
    -1.03
     neque
    -0.97
     面包
    -0.96
     namun
    -0.96
     τὴν
    -0.95
     triom
    -0.95
     religieuses
    -0.95
     profondeur
    -0.92
     afficher
    -0.92
    POSITIVE LOGITS
     evidence
    1.56
     documentation
    1.20
     documented
    1.19
     records
    1.17
     your
    1.17
     documenting
    1.16
     recorded
    1.15
     claim
    1.13
     document
    1.13
     recording
    1.11
    Activations Density 0.031%

    No Known Activations