© 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. 1634
    Prev
    Next
    INDEX
    Explanations

    alternation

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

    The neuron is detecting pieces of HTML character entity references (e.g. the “x27;” in “'”).

    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
     to
    -1.84
     with
    -1.70
    𖤣
    -1.68
     but
    -1.65
    }
    -1.57
     before
    -1.57
     it
    -1.55
     Médico
    -1.55
    );
    -1.54
     ὁ
    -1.52
    POSITIVE LOGITS
    或者
    1.77
    if
    1.70
    aliśmy
    1.63
    <bos>
    1.63
     espectaculares
    1.60
    we
    1.58
    chables
    1.57
     divertidos
    1.56
     altercation
    1.55
    currently
    1.54
    Activations Density 0.002%

    No Known Activations