© 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. 104053
    Prev
    Next
    INDEX
    Explanations

    you know or have seen

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

    The neuron fires on auxiliary‐“have” forms (especially “you’ve/you have” plus a past participle) marking perfect-aspect verb phrases.

    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
    ونی
    -1.14
    dirond
    -1.13
     felipe
    -1.13
     Infir
    -1.13
     amour
    -1.13
     COOKIE
    -1.12
    llten
    -1.12
     zorro
    -1.11
     XXIV
    -1.11
     quando
    -1.09
    POSITIVE LOGITS
     or
    1.75
     our
    1.74
     my
    1.69
     this
    1.63
     any
    1.59
     it
    1.54
     autre
    1.52
     another
    1.45
    </tfoot>
    1.41
    laient
    1.36
    Activations Density 0.068%

    No Known Activations