© 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. 10-GEMMASCOPE-RES-131K
    7. 58807
    Prev
    Next
    INDEX
    Explanations

    that then noun

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

    The neuron strongly activates on demonstrative/deictic words (e.g. “that,” “there,” “way”).

    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
    還會
    -1.41
     dwóch
    -1.36
    其他
    -1.34
     voorbeeld
    -1.34
     カウンター
    -1.32
     samme
    -1.31
     innych
    -1.30
     其他
    -1.30
     MAKES
    -1.30
    齮
    -1.30
    POSITIVE LOGITS
     some
    2.05
     if
    1.49
     as
    1.45
     primarily
    1.43
     Some
    1.42
     incorporating
    1.32
     instead
    1.31
     most
    1.31
     mostly
    1.31
     while
    1.28
    Activations Density 0.053%

    No Known Activations