© 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. 22-GEMMASCOPE-RES-131K
    4. 347
    Prev
    Next
    INDEX
    Explanations

    table formatting symbols

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

    This neuron fires for isolated, capitalized label or heading tokens (e.g. column or section titles like “field,” “operation,” “Version,” “Description”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     улыб
    -1.03
    ophysical
    -0.96
     його
    -0.96
     vedno
    -0.96
     שלנו
    -0.96
    /
    -0.95
    蜀
    -0.94
    куда
    -0.93
     ambition
    -0.93
    onus
    -0.93
    POSITIVE LOGITS
     verschied
    1.08
     martens
    1.08
    owi
    1.05
    oczes
    1.05
    ły
    1.05
     oogen
    1.04
     alfo
    1.04
     muft
    1.03
     дет
    1.02
    issima
    1.02
    Activations Density 0.020%

    No Known Activations