© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 104379
    Prev
    Next
    INDEX
    Explanations

    raw starts

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

    The neuron strongly fires on the subword token “ra” whenever it appears 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_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
    </h3>
    -3.53
    </strong>
    -3.13
    in
    -2.89
    榈
    -2.84
    to
    -2.84
    with
    -2.81
    也会
    -2.73
    You
    -2.67
    They
    -2.67
    It
    -2.66
    POSITIVE LOGITS
     myriad
    2.56
     dobry
    2.52
     ajouter
    2.41
    jenige
    2.38
    ᥀
    2.28
     then
    2.25
     juncture
    2.23
     nadru
    2.22
     lonely
    2.20
     koniec
    2.19
    Activations Density 0.002%

    No Known Activations