© 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. 75678
    Prev
    Next
    INDEX
    Explanations

    minus and excluding

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

    The neuron activates on occurrences of the word “minus” used as a subtraction sign or negative marker.

    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
    .”
    -3.14
    derdag
    -2.69
     countless
    -2.61
     unrelenting
    -2.44
    :”
    -2.42
     undeniably
    -2.36
    indelijk
    -2.28
    —”
    -2.25
     have
    -2.25
     initiatives
    -2.23
    POSITIVE LOGITS
    菒
    2.77
    絰
    2.75
    耵
    2.64
    ating
    2.58
    cule
    2.55
    脔
    2.53
    卝
    2.52
    鉏
    2.52
    蔟
    2.52
    it
    2.47
    Activations Density 0.004%

    No Known Activations