© 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. 20539
    Prev
    Next
    INDEX
    Explanations

    verb-ing noun phrase

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

    The neuron flags gerund (-ing) verbs that describe procedural or methodological actions (e.g. using, measuring, comparing) in technical/scientific text.

    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
     beiden
    -1.72
    breviations
    -1.71
     coreano
    -1.55
     liknande
    -1.49
    amente
    -1.49
     impressive
    -1.45
    ('=
    -1.45
    berian
    -1.41
     in
    -1.41
     was
    -1.41
    POSITIVE LOGITS
    stości
    1.38
     فقال
    1.24
     カル
    1.23
     célè
    1.22
    rítica
    1.22
     libremente
    1.21
     méd
    1.20
    йти
    1.17
    小声
    1.17
     Você
    1.16
    Activations Density 0.246%

    No Known Activations