© 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. 68316
    Prev
    Next
    INDEX
    Explanations

    French and German words

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

    The neuron is highly activated by numeric tokens (integers or decimals), i.e. it detects 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_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
     A
    -1.11
    this
    -1.09
    so
    -1.04
    des
    -1.02
    co
    -1.01
     d
    -1.01
     des
    -1.00
    '
    -0.98
    be
    -0.98
     It
    -0.98
    POSITIVE LOGITS
     tromper
    1.35
     farine
    1.29
    Dieu
    1.27
    selbst
    1.24
    nicht
    1.20
     magasiner
    1.20
     empêcher
    1.17
    können
    1.15
     digitalen
    1.14
    zurück
    1.13
    Activations Density 0.019%

    No Known Activations