© 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. 10-GEMMASCOPE-RES-131K
    7. 61056
    Prev
    Next
    INDEX
    Explanations

    attribute assignments

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

    The neuron activates on numeric literals (tokens consisting of numbers).

    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
    weren
    -2.53
     ì
    -2.44
    逦
    -2.36
    ikusbot
    -2.31
    啻
    -2.25
    wouldn
    -2.25
     »)
    -2.23
     ọ
    -2.22
    cuz
    -2.19
    esterday
    -2.17
    POSITIVE LOGITS
    s
    3.44
    ">
    2.34
    section
    2.34
    leq
    2.19
     increase
    2.16
    да
    2.13
     innere
    2.05
    т
    2.05
    寵
    2.02
    ="
    2.00
    Activations Density 0.002%

    No Known Activations