© 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. 42110
    Prev
    Next
    INDEX
    Explanations

    bringing someone or something along

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

    The neuron is highly sensitive to prepositions (especially “along,” “with,” “behind,” etc.) that signal accompaniment or spatial/relational context.

    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
     acogedor
    -1.66
     RSITY
    -1.53
     that
    -1.53
    NÍ
    -1.43
     Belgische
    -1.41
    
    -1.41
    libatkan
    -1.40
     enemigo
    -1.38
     prêts
    -1.36
     auront
    -1.35
    POSITIVE LOGITS
     in
    1.76
     to
    1.66
     when
    1.55
     Only
    1.49
     for
    1.45
     just
    1.44
     On
    1.39
     if
    1.39
     on
    1.30
    祈
    1.25
    Activations Density 0.014%

    No Known Activations