© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 128940
    Prev
    Next
    INDEX
    Explanations

    volcanoes, volleyball, volume

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

    This neuron activates on any token containing the substring “vol,” such as “vol,” “Vol,” “Volcano,” or “volleyball.”

    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
    
    -2.59
    which
    -2.55
     disfruta
    -2.50
     vollständ
    -2.38
     ―
    -2.31
     zusätz
    -2.28
     generando
    -2.28
    
    -2.28
    仃
    -2.27
    
    -2.25
    POSITIVE LOGITS
    vol
    3.08
    𓐍
    3.02
     {
    2.89
    ꪔ
    2.84
    𓆸
    2.58
     =>
    2.55
     was
    2.55
    𓆜
    2.53
    "?
    2.52
    the
    2.45
    Activations Density 0.013%

    No Known Activations