© 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. 26553
    Prev
    Next
    INDEX
    Explanations

    specific noun pairings

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

    The neuron responds to long, multi-syllable words—especially formal or technical terms.

    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
     FUCK
    -1.29
    Jika
    -1.29
    日時
    -1.28
     “…
    -1.27
    ıza
    -1.25
     Jika
    -1.23
     bArr
    -1.23
    امین
    -1.18
    Är
    -1.17
     Kristus
    -1.17
    POSITIVE LOGITS
     maybe
    2.00
     things
    1.95
     kind
    1.94
     stuff
    1.91
     really
    1.91
     thing
    1.87
     you
    1.80
     we
    1.79
     [
    1.76
     what
    1.76
    Activations Density 0.024%

    No Known Activations