© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65878
    Prev
    Next
    INDEX
    Explanations

    initialisms and proper nouns

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

    The neuron fires on the initial few letters of longer capitalized words, i.e. on the prefixes of proper nouns.

    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
    酲
    -2.61
    ти
    -2.52
     öffentliche
    -2.50
    鋮
    -2.50
    kwür
    -2.38
    镝
    -2.36
     offizielle
    -2.30
     gustaMe
    -2.27
     This
    -2.25
     größere
    -2.23
    POSITIVE LOGITS
    </strong>
    3.25
    u
    2.92
    and
    2.88
     hebben
    2.70
    </h3>
    2.61
    d
    2.50
    s
    2.48
    al
    2.47
    盉
    2.41
    y
    2.36
    Activations Density 0.007%

    No Known Activations