© 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. 130690
    Prev
    Next
    INDEX
    Explanations

    names or prefixes followed by specific words

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

    The neuron selectively activates on the initial subword pieces of longer, less common proper names 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_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
     for
    -3.44
     so
    -3.34
     has
    -3.17
     after
    -2.77
     what
    -2.75
     but
    -2.61
     before
    -2.56
     because
    -2.56
     just
    -2.56
     and
    -2.56
    POSITIVE LOGITS
     trám
    2.92
     refroid
    2.59
     normalt
    2.56
    粜
    2.56
    躓
    2.47
     rafraî
    2.45
     régal
    2.41
     alltid
    2.31
     corriger
    2.30
     eerste
    2.28
    Activations Density 0.704%

    No Known Activations