© 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. 25302
    Prev
    Next
    INDEX
    Explanations

    Unknown database

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

    The neuron responds to occurrences of the substring “hyper” (as in “hypertensive,” “hyper-​tensive,” 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
    бовать
    -0.86
     raspberries
    -0.79
     workman
    -0.79
     tristesse
    -0.79
     risque
    -0.78
    uncias
    -0.77
     exercer
    -0.75
    levé
    -0.75
    自行
    -0.75
    arie
    -0.74
    POSITIVE LOGITS
     later
    0.89
    sự
    0.86
     também
    0.83
     also
    0.82
    eseorang
    0.81
     també
    0.80
    Hva
    0.77
    umbrella
    0.77
    đại
    0.76
    ᆨ
    0.76
    Activations Density 0.039%

    No Known Activations