© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 48474
    Prev
    Next
    INDEX
    Explanations

    can do something

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

    The neuron fires on Russian modal verbs (especially “могу/могут” and similar) indicating ability or permission.

    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
    WRENCE
    -2.42
    riezmann
    -2.30
    licitations
    -2.23
    elerik
    -2.06
    inerary
    -2.03
    ysteine
    -2.03
     hierzu
    -2.02
    の魅力
    -2.02
    ícias
    -1.98
    勐
    -1.98
    POSITIVE LOGITS
    ↵↵
    2.44
    ihnach
    1.96
     如果
    1.89
     If
    1.88
    T
    1.84
     två
    1.84
     be
    1.83
    </td>
    1.80
     It
    1.77
    H
    1.75
    Activations Density 0.001%

    No Known Activations