© 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. 20293
    Prev
    Next
    INDEX
    Explanations

    die game

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

    This neuron fires on the words in the question’s title or heading (the slug‐style or title‐case phrases at the top of each post).

    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
     kristal
    -1.52
     biblia
    -1.49
     legenda
    -1.48
     peruan
    -1.41
     dizi
    -1.40
     pelicula
    -1.39
     loh
    -1.38
     tava
    -1.38
     symbole
    -1.38
     tempel
    -1.38
    POSITIVE LOGITS
     first
    1.42
     also
    1.27
     new
    1.22
     proszę
    1.19
     everyday
    1.18
     począ
    1.17
     además
    1.17
    selbe
    1.16
     be
    1.13
     extreme
    1.13
    Activations Density 0.107%

    No Known Activations