© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130847
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on any token containing the substring “Tom” (e.g. Tom, Tomcat, Thomas).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Tom and Thomas names, titles, or compound words

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ?"
    -2.42
    也
    -2.42
    on
    -2.41
    g
    -2.31
    ！
    -2.14
    是最
    -2.13
     After
    -2.11
    there
    -2.09
    we
    -2.05
    s
    -2.02
    POSITIVE LOGITS
    釤
    3.30
    𓆟
    3.17
    翃
    2.67
     we
    2.61
    ᛬
    2.42
    珦
    2.41
    を読む
    2.33
     efekty
    2.28
    ditionally
    2.28
     manchen
    2.27
    Activations Density 0.042%

    No Known Activations