© 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. 85231
    Prev
    Next
    INDEX
    Explanations

    Ball, SubDir, Julian von, Суть

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

    This neuron fires on tokens beginning with the letter sequence “su” (in any case).

    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
     leaſt
    -2.81
     dritte
    -2.67
     greateſt
    -2.61
     of
    -2.58
    乀
    -2.56
    Ꮞ
    -2.48
    赑
    -2.47
    presumably
    -2.45
    ではまた
    -2.39
    usually
    -2.38
    POSITIVE LOGITS
    *
    2.89
     {
    2.81
    \
    2.70
    i
    2.61
     extraordinarily
    2.58
     decipher
    2.45
     vehemently
    2.44
     meticulously
    2.41
    いえば
    2.41
     awfully
    2.41
    Activations Density 0.000%

    No Known Activations