© 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. 26081
    Prev
    Next
    INDEX
    Explanations

    various types of

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

    The neuron detects generic category-introducing phrases that signal classification (e.g. “types of,” “kinds of”).

    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
    ressing
    -1.62
    但却
    -1.45
    丝毫
    -1.42
    察觉
    -1.41
    inator
    -1.41
     inalámbrico
    -1.41
    桫
    -1.40
    phere
    -1.39
    ture
    -1.38
     ét
    -1.38
    POSITIVE LOGITS
     things
    1.72
     مختلف
    1.65
     types
    1.60
     هستند
    1.45
    Beskrivning
    1.41
    いかがでしょうか
    1.38
    tarta
    1.38
     situations
    1.38
    也會
    1.38
     different
    1.36
    Activations Density 0.028%

    No Known Activations