© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71847
    Prev
    Next
    INDEX
    Explanations

    language construction and code

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

    This neuron responds to numeric tokens (integer number strings) 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
     people
    -1.57
     just
    -1.38
     серый
    -1.35
     how
    -1.30
     not
    -1.30
     their
    -1.27
     several
    -1.24
     malu
    -1.23
    日閲覧
    -1.22
     why
    -1.19
    POSITIVE LOGITS
    了过来
    1.21
    了出来
    1.17
     echte
    1.16
    filed
    1.12
     and
    1.11
     permett
    1.11
    intégr
    1.09
    she
    1.08
    readFileSync
    1.08
     demnach
    1.07
    Activations Density 0.008%

    No Known Activations