© 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. 34-GEMMASCOPE-RES-131K
    7. 47825
    Prev
    Next
    INDEX
    Explanations

    legal, listing, superficial, new, goal, question, HN

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

    The neuron fires on the all-caps headings and key label tokens that mark section titles in real-estate listing 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
    😜
    -0.78
     getRandom
    -0.76
     perg
    -0.75
     保育
    -0.74
    😤
    -0.73
    🤤
    -0.72
     PUBL
    -0.71
    
    -0.71
    🤑
    -0.69
    😈
    -0.69
    POSITIVE LOGITS
    друг
    0.79
     listing
    0.76
    Essa
    0.75
     Wainwright
    0.73
     vollständ
    0.73
     Biaya
    0.72
     động
    0.72
    pleaños
    0.72
    ặp
    0.71
     geisti
    0.71
    Activations Density 0.005%

    No Known Activations