© 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. 22-GEMMASCOPE-RES-131K
    7. 73816
    Prev
    Next
    INDEX
    Explanations

    foreign language and common words

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

    The neuron detects tokens that signal the start of a new segment (e.g. interjection or header tokens at the beginnings of snippets).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    こいつ
    -1.25
     kadang
    -1.22
     pantalón
    -1.21
     in
    -1.11
     ज्ञात
    -1.10
    なかな
    -1.09
    lielmo
    -1.09
     recently
    -1.08
     for
    -1.07
     December
    -1.06
    POSITIVE LOGITS
     finally
    1.46
     again
    1.26
    "}";
    1.25
     sämtliche
    1.24
     I
    1.23
    mainAxisSize
    1.19
    やはり
    1.18
     descobri
    1.17
    uable
    1.16
     esportes
    1.16
    Activations Density 0.008%

    No Known Activations