© 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. 41643
    Prev
    Next
    INDEX
    Explanations

    policy terms and roles

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

    The neuron fires on key words that denote formal programs, initiatives, or organizational titles (e.g. “Future,” “Innovation,” “Fellowship,” “Forum,” “Preparedness”).

    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
     mennesker
    -1.48
    gebraucht
    -1.47
     grij
    -1.44
     apartament
    -1.41
    geladen
    -1.41
     medlemmer
    -1.41
     magasiner
    -1.39
    kommt
    -1.38
     normaal
    -1.38
     illustrationer
    -1.33
    POSITIVE LOGITS
     o
    1.42
     show
    1.40
     t
    1.38
     find
    1.30
     never
    1.30
    ürgen
    1.30
    "-
    1.28
     .
    1.28
     including
    1.27
    在其
    1.26
    Activations Density 0.189%

    No Known Activations