© 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. 67493
    Prev
    Next
    INDEX
    Explanations

    может быть, возможно, Dyspnea, Site Navigation, Q:

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

    The neuron fires on Russian modal‐auxiliary verbs expressing possibility (e.g. может, быть, могут, будет).

    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
    鮡
    -2.75
    翃
    -2.69
     komfor
    -2.64
    粜
    -2.61
    饸
    -2.59
     excès
    -2.52
    馓
    -2.48
     kompakt
    -2.41
    糨
    -2.41
     konfig
    -2.39
    POSITIVE LOGITS
     It
    3.31
    </b>
    3.05
     может
    3.02
     a
    2.86
     With
    2.63
     is
    2.45
    》，
    2.44
     D
    2.41
    </h2>
    2.34
    </h4>
    2.33
    Activations Density 0.001%

    No Known Activations