© 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. 10-GEMMASCOPE-RES-131K
    4. 68885
    Prev
    Next
    INDEX
    Explanations

    killer spy, killer whales, killer instinct

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

    The neuron activates whenever it sees the word “killer” (in any casing or as part of a compound).

    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
    '
    -2.53
    b
    -2.50
    t
    -2.47
    ?
    -2.45
    苼
    -2.34
    Зно
    -2.25
     flesta
    -2.20
     littered
    -2.19
     eſſ
    -2.17
     argentinos
    -2.13
    POSITIVE LOGITS
    .
    2.52
     furthermore
    2.45
    );
    2.27
    idlertid
    2.25
     pôr
    2.23
    胧
    2.23
     grabar
    2.19
    なども
    2.14
     "-"
    2.09
     también
    2.06
    Activations Density 0.003%

    No Known Activations