© 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. 72841
    Prev
    Next
    INDEX
    Explanations

    German, Irish, Japanese words

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

    This neuron responds to rare, multi-syllable tokens—especially unusual proper names and specialized terms.

    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
     travaillent
    -2.23
    
    -2.20
    mpagne
    -2.17
    СТО
    -2.17
    СИ
    -2.16
     prennent
    -2.14
    苤
    -2.11
     queridos
    -2.05
    dchen
    -2.05
     Autó
    -2.05
    POSITIVE LOGITS
    <td>
    2.72
    s
    2.67
    ではでは
    2.67
    If
    2.52
    p
    2.48
    k
    2.47
    さてさて
    2.42
     this
    2.34
    An
    2.33
    f
    2.33
    Activations Density 0.065%

    No Known Activations