© 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. 30918
    Prev
    Next
    INDEX
    Explanations

    starts phrases like Q: or For

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

    The neuron strongly activates on named entities or titles (proper names of things, like product names, class names, or quoted “titles”).

    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
     wiedzieć
    -0.84
    Authorization
    -0.81
    حاد
    -0.78
     Merchants
    -0.77
    érat
    -0.76
    ص
    -0.75
    Sv
    -0.75
    Meet
    -0.75
    ǰ
    -0.74
    Ү
    -0.73
    POSITIVE LOGITS
     übrig
    0.90
     Evo
    0.88
    guantes
    0.84
    熟悉的
    0.82
     müm
    0.81
    が残
    0.80
    ाम
    0.78
     생각
    0.76
    LogService
    0.76
    Nour
    0.75
    Activations Density 0.000%

    No Known Activations