© 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. 34406
    Prev
    Next
    INDEX
    Explanations

    close relationships

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

    The neuron strongly activates on words that denote social or group affiliation (e.g. “brothers,” “sisters,” “friends,” “colleagues,” “fellow,” “humans,” 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
     Daarnaast
    -1.14
    Недоліки
    -1.12
    ⡦
    -1.09
     chiesa
    -1.03
    小结
    -1.01
     após
    -1.00
     Américas
    -0.99
     überzeugt
    -0.98
     maisha
    -0.95
     nghiên
    -0.94
    POSITIVE LOGITS
     from
    1.69
     who
    1.34
     across
    1.09
     in
    1.07
    нные
    1.01
    we
    0.96
    。』
    0.94
    especially
    0.94
    whose
    0.92
    琑
    0.91
    Activations Density 0.023%

    No Known Activations