© 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. 22-GEMMASCOPE-RES-131K
    7. 25441
    Prev
    Next
    INDEX
    Explanations

    pop-up, populating, or launch

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

    The neuron consistently activates on the letter sequence “pop” (i.e. words beginning with “pop,” like pop-up, populate, popculture, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     have
    -2.56
    ми
    -2.44
     to
    -2.27
    .”
    -2.25
     rarely
    -2.19
    啬
    -2.14
    vantaged
    -2.14
    >
    -2.13
     Siehe
    -2.08
    歌唱
    -2.05
    POSITIVE LOGITS
    佧
    2.88
    utiva
    2.72
    ꜆
    2.70
    erintah
    2.55
     chrétien
    2.47
    
    2.47
    ecie
    2.39
    iapkan
    2.39
    Jangan
    2.38
     использовании
    2.36
    Activations Density 0.013%

    No Known Activations