© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113939
    Prev
    Next
    INDEX
    Explanations

    shed new light or wreak havoc

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

    The neuron fires most strongly on action verbs (especially transitive verbs in their base or infinitive form) that describe causing or exerting some effect.

    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
    鸶
    -1.98
    ᯾
    -1.92
    鳚
    -1.85
    桖
    -1.82
    廔
    -1.78
    𓋼
    -1.77
    ⽌
    -1.77
    ÉE
    -1.75
     noël
    -1.73
     Worten
    -1.73
    POSITIVE LOGITS
    '
    2.47
     “
    1.90
    ”
    1.80
    ↵↵
    1.78
    --
    1.77
     a
    1.74
     ønsker
    1.70
    “
    1.67
    <bos>
    1.66
     kanskje
    1.59
    Activations Density 0.071%

    No Known Activations