© 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. 73751
    Prev
    Next
    INDEX
    Explanations

    suppose or assume conditional statements

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

    The neuron is strongly activated by numeric expressions (digits, decimals, percentages, and related punctuation) in the text.

    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
     certamente
    -1.47
     quando
    -1.46
    戗
    -1.44
     künftig
    -1.41
     grandiose
    -1.41
    ᠂
    -1.39
     notwendigen
    -1.36
    ributo
    -1.35
     biblia
    -1.34
    likon
    -1.33
    POSITIVE LOGITS
     that
    2.11
     and
    1.90
     one
    1.59
    Also
    1.56
     at
    1.45
     four
    1.42
     only
    1.42
     TWO
    1.41
     JUST
    1.37
     around
    1.35
    Activations Density 0.093%

    No Known Activations