© 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. 73285
    Prev
    Next
    INDEX
    Explanations

    software.target.all.my.framework.iOS

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

    The neuron is highly sensitive to specialized technical or scientific terminology—that is, domain‐specific jargon and keywords.

    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
     MARKETS
    -0.92
     Utf
    -0.85
     медицин
    -0.85
     Angew
    -0.84
     jenem
    -0.83
     Protobuf
    -0.82
    binant
    -0.81
     éco
    -0.81
    ceuticals
    -0.81
    出去
    -0.80
    POSITIVE LOGITS
    Then
    1.00
    циям
    0.89
    cuatro
    0.82
    忱
    0.82
    There
    0.81
    œuvres
    0.80
     ͡°
    0.79
    If
    0.79
    Other
    0.78
    ございます
    0.78
    Activations Density 0.114%

    No Known Activations