© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 86895
    Prev
    Next
    INDEX
    Explanations

    GraphicalEditor se Graphicaleditor user GraphicalEditor GraphicalEditor GraphicalEditor se GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor GraphicalEditor

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

    The neuron activates on double-asterisk markers (e.g. “**”) used for bold/emphasis formatting.

    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
    珢
    -3.11
    geſ
    -2.92
    砗
    -2.88
    茕
    -2.86
    戆
    -2.80
    亖
    -2.77
    薆
    -2.72
    蚍
    -2.69
    ⬈
    -2.64
    ቺ
    -2.58
    POSITIVE LOGITS
     Also
    3.48
     Until
    3.20
     Besides
    3.19
     Whereas
    3.17
     Although
    3.16
     Therefore
    3.11
     Certainly
    3.08
     Presumably
    3.06
     Except
    3.03
     Another
    3.02
    Activations Density 0.045%

    No Known Activations