© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26775
    Prev
    Next
    INDEX
    Explanations

    structured data keys

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

    The neuron activates on numeric tokens—i.e. it’s looking for numbers (numeric literals) 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
     régal
    -1.99
     siè
    -1.76
     célè
    -1.73
     estadounid
    -1.70
     pittores
    -1.66
     oublier
    -1.63
     baumwolle
    -1.63
     花纹
    -1.62
     rése
    -1.60
     inond
    -1.60
    POSITIVE LOGITS
     q
    1.74
     L
    1.68
     p
    1.62
    L
    1.62
     t
    1.62
     h
    1.62
     f
    1.61
     took
    1.58
     F
    1.57
     C
    1.56
    Activations Density 0.008%

    No Known Activations