© 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. 34-GEMMASCOPE-RES-131K
    7. 7166
    Prev
    Next
    INDEX
    Explanations

    physics and mathematics departments

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

    The neuron is detecting academic affiliation phrases—department or institute names (e.g. “Department of Physics,” “Institute of Oncology,” etc.) in author/address sections.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     tecnica
    -0.91
    }}_{\
    -0.90
     filosofia
    -0.89
     everything
    -0.88
     dupa
    -0.87
     while
    -0.83
     comuni
    -0.82
     consentimento
    -0.81
     risultato
    -0.81
     caratteri
    -0.81
    POSITIVE LOGITS
     department
    1.25
    Faculty
    1.16
     faculty
    1.13
     Faculty
    1.06
    Especially
    0.99
     and
    0.95
    และ
    0.93
    faculty
    0.93
    ensión
    0.91
     surtout
    0.85
    Activations Density 0.042%

    No Known Activations