© 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. 10-GEMMASCOPE-RES-131K
    7. 41204
    Prev
    Next
    INDEX
    Explanations

    Roman numerals, numbers, and references

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

    The neuron consistently activates on standalone numeric tokens (especially single digits and list‐style numbers), i.e. it detects occurrences of numbers.

    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
     conector
    -1.54
     Departamento
    -1.32
    yoruz
    -1.31
     satte
    -1.28
     היא
    -1.27
     although
    -1.25
     bloss
    -1.23
    urlopen
    -1.23
    getCustomer
    -1.22
     surg
    -1.21
    POSITIVE LOGITS
     gegenwär
    1.80
     terá
    1.67
    various
    1.66
     żad
    1.64
     vorüber
    1.63
    ↵
    1.58
     focuses
    1.58
     außerord
    1.56
    从来
    1.54
     besonderen
    1.54
    Activations Density 0.029%

    No Known Activations