© 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. 90861
    Prev
    Next
    INDEX
    Explanations

    with/in + action/purpose

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

    The neuron fires strongly on verbs that denote providing assistance or help (e.g. “help,” “assist,” “aid”) especially when paired with “with” or “in.”

    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
    极其
    -1.63
     elevador
    -1.59
     reluctance
    -1.59
     comprometido
    -1.59
    ’，
    -1.57
     religiosas
    -1.54
     firmar
    -1.52
     überzeugt
    -1.52
     excelentes
    -1.52
     domés
    -1.52
    POSITIVE LOGITS
    );
    1.92
     =
    1.83
    T
    1.74
    J
    1.67
    tanleria
    1.62
    P
    1.59
    U
    1.57
     like
    1.54
     also
    1.54
    C
    1.51
    Activations Density 0.011%

    No Known Activations