© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 73569
    Prev
    Next
    INDEX
    Explanations

    special + role adjective

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

    The neuron is specialized to detect the word “Special” (as in headings or titles).

    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
     Stew
    -0.76
     Tast
    -0.75
    共有
    -0.75
     tentacles
    -0.70
     tuner
    -0.68
     Maro
    -0.68
     andes
    -0.67
    owanych
    -0.67
    ECTO
    -0.66
     bidder
    -0.66
    POSITIVE LOGITS
     populations
    1.04
    TY
    0.90
     projects
    0.79
    ty
    0.78
    vette
    0.75
    Gardner
    0.75
    нее
    0.75
     studies
    0.75
    Dip
    0.75
    Trọng
    0.75
    Activations Density 0.034%

    No Known Activations