© 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. 33658
    Prev
    Next
    INDEX
    Explanations

    abstract concepts and outcomes

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

    The neuron flags high‐level academic or review section words—abstract topic labels like “history,” “characteristics,” “mechanisms,” “physiology,” 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_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
     or
    -1.14
    With
    -1.02
    以下の
    -1.01
    ........
    -0.99
    !!!!
    -0.98
    Combination
    -0.97
     ausreich
    -0.97
     more
    -0.97
    ↵
    -0.97
     quotient
    -0.95
    POSITIVE LOGITS
     and
    1.72
     of
    1.55
    '';
    1.28
     italie
    1.23
     Offisielt
    1.22
    erobic
    1.20
     parachoque
    1.20
     résister
    1.19
     inox
    1.19
     män
    1.16
    Activations Density 0.048%

    No Known Activations