© 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. 84257
    Prev
    Next
    INDEX
    Explanations

    family

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

    The neuron strongly activates on the word “family” (and its close inflection “families”), marking mentions of a group or community affiliation.

    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
     dazzling
    -1.73
     striving
    -1.55
     or
    -1.55
     stately
    -1.47
     Perhaps
    -1.47
     enticing
    -1.45
     characteristics
    -1.43
    bumptech
    -1.43
     unrelenting
    -1.42
     prevent
    -1.41
    POSITIVE LOGITS
    圌
    1.81
     remarkably
    1.61
    姈
    1.59
     verdura
    1.58
    ￬
    1.56
     raiſ
    1.48
     produisent
    1.48
     zoude
    1.48
     vastly
    1.48
     creat
    1.47
    Activations Density 0.020%

    No Known Activations