© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 22120
    Prev
    Next
    INDEX
    Explanations

    at a glance

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

    The neuron flags individual tokens that participate in common multiword collocations or fixed expressions (e.g. basket of, at my fingertips, wreak havoc).

    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
    3
    -1.71
    tabilité
    -1.66
     others
    -1.66
    0
    -1.63
     allure
    -1.59
    5
    -1.59
    6
    -1.50
    7
    -1.47
    os
    -1.42
     و
    -1.42
    POSITIVE LOGITS
    ılması
    1.45
     solicitó
    1.45
    縂
    1.40
     territ
    1.37
    untuk
    1.33
    
    1.33
    gilet
    1.31
     Affaires
    1.30
    Gobierno
    1.28
     الفلسط
    1.27
    Activations Density 0.077%

    No Known Activations