© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 48273
    Prev
    Next
    INDEX
    Explanations

    citations and references

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

    The neuron selectively fires on multi‐digit numeric tokens (e.g. page numbers, years, table indices—in short, any numeric strings).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     or
    -1.34
     like
    -1.31
     despite
    -1.26
     such
    -1.20
     then
    -1.17
     noteworthy
    -1.16
     which
    -1.07
     just
    -1.06
     notable
    -1.05
     compelling
    -1.05
    POSITIVE LOGITS
     verschillende
    1.59
     &,
    1.32
     verschiedenen
    1.15
    ³,
    1.15
     vielen
    1.14
     %,
    1.14
    ​​​
    1.13
    ²,
    1.13
    **,
    1.13
    Again
    1.09
    Activations Density 0.049%

    No Known Activations