© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99594
    Prev
    Next
    INDEX
    Explanations

    logos and emblems

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

    The neuron activates on words naming heraldic charges or shield elements (i.e. terms used in coat-of-arms descriptions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     propOrder
    -0.90
    persky
    -0.84
    ghurt
    -0.84
    unno
    -0.83
    ੫
    -0.82
    给大家
    -0.81
    absolutely
    -0.80
    chaub
    -0.79
    comy
    -0.79
    rename
    -0.78
    POSITIVE LOGITS
     fleur
    1.36
     crescent
    1.24
     logo
    1.18
     red
    1.17
     letter
    1.14
     skull
    1.12
     triangle
    1.11
     star
    1.09
     circle
    1.07
     oval
    1.03
    Activations Density 0.049%

    No Known Activations