© 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. 61364
    Prev
    Next
    INDEX
    Explanations

    sensational,UserController, orden, Seleccion

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

    The neuron strongly activates on unusually long words (tokens with many characters), regardless of language.

    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
     were
    -3.06
    *
    -2.70
     surpass
    -2.45
     Could
    -2.38
    ي
    -2.36
    sonaro
    -2.30
     advent
    -2.27
     You
    -2.27
     menschliche
    -2.25
     Would
    -2.23
    POSITIVE LOGITS
    ing
    3.69
    ro
    2.52
    an
    2.45
    );
    2.44
    ので
    2.31
    殽
    2.25
    and
    2.22
     SOCK
    2.22
    他也
    2.20
     ストレッチ
    2.19
    Activations Density 0.002%

    No Known Activations