© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 100302
    Prev
    Next
    INDEX
    Explanations

    richness and substance

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

    The neuron activates on words and phrases that convey adding substance or making something more substantial (e.g. “flesh out,” “rich,” “robust,” “meat,” “beef up”).

    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
    旆
    -1.41
    閦
    -1.35
    )";
    
    -1.32
     marte
    -1.31
    vų
    -1.30
    )]
    
    -1.28
     paco
    -1.27
     vorigen
    -1.26
    */
    
    
    -1.25
     eſſ
    -1.25
    POSITIVE LOGITS
    外人
    1.27
     isolada
    1.23
    ómico
    1.23
     которая
    1.22
    了他的
    1.20
     \
    1.20
     and
    1.18
    عنی
    1.18
     is
    1.13
    fillText
    1.13
    Activations Density 0.054%

    No Known Activations