© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85749
    Prev
    Next
    INDEX
    Explanations

    uncivilized, primitive, savage

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

    The neuron strongly activates on the word “savage” (or very close morphological variants of it).

    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
    塍
    -2.27
    穑
    -2.19
    幛
    -2.09
    奩
    -2.06
    身边的
    -2.06
     trucos
    -2.03
    ᠁
    -2.02
    扨
    -2.00
     BRITAIN
    -1.99
    跸
    -1.93
    POSITIVE LOGITS
     are
    2.77
     While
    2.34
     Despite
    2.14
     Since
    2.14
     With
    2.13
     Ideally
    2.13
     During
    2.09
     Although
    2.06
    While
    2.03
     После
    2.02
    Activations Density 0.009%

    No Known Activations