© 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. 9679
    Prev
    Next
    INDEX
    Explanations

    because last

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

    The neuron fires strongly on numeric tokens (especially scores, statistics, and other numbers), effectively detecting numbers in the text.

    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
     of
    -1.85
     also
    -1.73
     årene
    -1.59
     rêver
    -1.44
     millioner
    -1.41
     måneder
    -1.31
    ślę
    -1.29
    тое
    -1.27
     hjemmes
    -1.24
    jähriger
    -1.24
    POSITIVE LOGITS
     conocemos
    1.20
     Porque
    1.19
     gdyż
    1.17
     
    1.17
    becca
    1.17
    ꧂
    1.16
     ponieważ
    1.16
    ",
    1.15
    ><><
    1.14
     desactivar
    1.14
    Activations Density 0.143%

    No Known Activations