© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65517
    Prev
    Next
    INDEX
    Explanations

    same or similar

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

    The neuron fires on sentence boundary markers—e.g. end-of-sentence periods (and the first words of new sentences like “No,” “If,” or “It”).

    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
     this
    -0.94
    ographi
    -0.93
     AGRIC
    -0.92
     jawab
    -0.89
     that
    -0.87
     it
    -0.86
    astu
    -0.85
    hemi
    -0.85
     ciertamente
    -0.85
     soportar
    -0.85
    POSITIVE LOGITS
     ebenfalls
    1.66
     likewise
    1.42
     Similarly
    1.34
     ditto
    1.31
     тоже
    1.29
     similarly
    1.29
    Similarly
    1.23
    同樣
    1.16
     Likewise
    1.16
    同様
    1.16
    Activations Density 0.029%

    No Known Activations