© 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. 15546
    Prev
    Next
    INDEX
    Explanations

    wanna followed by an action verb

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

    The neuron primarily detects the informal contraction “wanna.”

    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
     \
    -2.98
    葑
    -2.72
     necesarias
    -2.70
     okazji
    -2.58
     Strukt
    -2.55
     kiedy
    -2.45
    戗
    -2.34
    鉝
    -2.31
    푣
    -2.31
    有助于
    -2.28
    POSITIVE LOGITS
    -
    2.42
     préféré
    2.38
     Many
    2.17
    笮
    2.16
    食物
    2.14
    out
    2.06
    </h2>
    2.05
    This
    2.03
    硨
    2.03
    er
    2.02
    Activations Density 0.002%

    No Known Activations