© 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. 129216
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on capitalized product, technology, or topic names (e.g. “Salesforce,” “Drum Channel,” “PDF,” “lua,” etc.).

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

    noun followed by related descriptor

    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
    Nazionale
    -1.73
     ujar
    -1.61
     réjou
    -1.59
    تول
    -1.59
     mochilas
    -1.58
     kompeti
    -1.57
    zdo
    -1.56
     coincid
    -1.56
    bánh
    -1.56
    مقال
    -1.55
    POSITIVE LOGITS
     is
    1.80
    L
    1.70
    b
    1.66
    f
    1.63
    _
    1.59
    E
    1.58
    l
    1.55
    B
    1.52
    a
    1.48
    x
    1.48
    Activations Density 0.128%

    No Known Activations