© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 9032
    Prev
    Next
    INDEX
    Explanations

    detects common words and characters

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

    The neuron fires on short Serbo-Croatian auxiliary/“to be” verb forms (e.g. je, smo, su) and related Slavic inflections.

    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
    1
    -2.80
    Two
    -2.69
    An
    -2.56
    How
    -2.55
    When
    -2.50
    an
    -2.50
    One
    -2.48
    Recently
    -2.48
    ug
    -2.42
    也會
    -2.42
    POSITIVE LOGITS
    璝
    2.67
    眨眼
    2.59
     instala
    2.56
    mpagne
    2.53
     detecta
    2.44
     justamente
    2.41
    沇
    2.36
     navide
    2.31
    跺
    2.31
    reathe
    2.28
    Activations Density 0.001%

    No Known Activations