© 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. 22-GEMMASCOPE-RES-131K
    7. 73849
    Prev
    Next
    INDEX
    Explanations

    quoted phrases and titles

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

    The neuron fires strongly on the opening of quoted passages (i.e. the initial quotation marks and the first word inside a quotation).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     also
    -1.25
     también
    -1.13
    also
    -1.11
    すば
    -1.06
    死ぬ
    -1.04
    ه‌ی
    -1.01
     asimismo
    -1.01
    '/',
    -0.98
    繋が
    -0.97
     представители
    -0.97
    POSITIVE LOGITS
     to
    1.41
    _
    1.33
     simplifié
    1.33
     Juifs
    1.20
    imread
    1.19
    はなく
    1.13
     egentlig
    1.09
    neuf
    1.09
    новые
    1.08
     chiede
    1.07
    Activations Density 0.029%

    No Known Activations