© 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. 58767
    Prev
    Next
    INDEX
    Explanations

    wide, wider, widest

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

    The neuron strongly activates on extremely common short function words and very high‐frequency tokens (e.g. “the,” “in,” “out,” “go,” “stay”).

    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
     Ks
    -0.75
    MYSQL
    -0.75
    VIC
    -0.71
     Cian
    -0.71
    Tä
    -0.71
     aplicada
    -0.69
    ⅳ
    -0.69
     democracia
    -0.69
     armoured
    -0.69
     deforma
    -0.69
    POSITIVE LOGITS
     wide
    3.58
     Wide
    2.83
    Wide
    2.70
    wide
    2.55
     wider
    2.45
     WIDE
    2.31
    WIDE
    2.19
     широ
    1.98
     širo
    1.89
     widest
    1.81
    Activations Density 0.019%

    No Known Activations