© 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. 15183
    Prev
    Next
    INDEX
    Explanations

    leader / favorite / source / founder

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

    The neuron fires on evaluative adjectives and qualifiers that convey strong emphasis or certainty (e.g. clear, convincing, overwhelming, unanimous).

    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
     that
    -2.80
     to
    -1.96
     then
    -1.70
     can
    -1.68
     pengusaha
    -1.57
     pemuda
    -1.53
     ciertos
    -1.44
     Then
    -1.44
     (
    -1.42
    揶
    -1.41
    POSITIVE LOGITS
     нашем
    1.59
     leader
    1.52
     winner
    1.46
     title
    1.42
    ൩
    1.38
     vinos
    1.38
    linawan
    1.38
    糨
    1.36
     huevos
    1.35
     depoz
    1.35
    Activations Density 0.067%

    No Known Activations