© 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. 128759
    Prev
    Next
    INDEX
    Explanations

    The neuron is highly sensitive to multi-digit numeric tokens (i.e. sequences of digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    numbered lists, headings, or initialisms

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     anún
    -1.94
     温泉
    -1.84
     técn
    -1.71
     séjours
    -1.61
     écoles
    -1.61
    𖡼
    -1.59
    鹀
    -1.55
    dijual
    -1.52
     leçons
    -1.52
    龔
    -1.52
    POSITIVE LOGITS
    N
    2.53
    B
    2.52
    J
    2.44
    D
    2.34
    S
    2.33
    V
    2.25
    C
    2.20
    T
    2.19
    L
    2.14
    U
    2.14
    Activations Density 0.049%

    No Known Activations