© 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. 22-GEMMASCOPE-RES-131K
    7. 65931
    Prev
    Next
    INDEX
    Explanations

    future predictions or confirmations

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

    The neuron fires on mentions of tools or similar equipment (e.g. “tool,” “tools,” “sterilized tool,” “these…”).

    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
     récompense
    -0.96
    یزات
    -0.95
    をはじめ
    -0.95
     combines
    -0.93
    愛用
    -0.90
     sonre
    -0.90
    幼稚園
    -0.89
    sneaker
    -0.88
     Estamos
    -0.88
    <bos>
    -0.88
    POSITIVE LOGITS
    事も
    1.26
    為に
    1.20
     then
    1.05
    際には
    1.03
     dann
    1.02
     dieser
    0.98
     xii
    0.98
    うまい
    0.94
     xvii
    0.94
    際は
    0.93
    Activations Density 0.059%

    No Known Activations