© 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. 65045
    Prev
    Next
    INDEX
    Explanations

    specific instances or quantities

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

    The neuron strongly activates on common nouns in their plural form (i.e. words ending in “s” that denote multiple items).

    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
     waarschijnlijk
    -1.10
     noodzake
    -0.94
     zicht
    -0.91
    Couleur
    -0.90
    fcntl
    -0.89
     همیشه
    -0.88
     مجلسی
    -0.88
     ontvangen
    -0.88
    nými
    -0.87
    ḗ
    -0.87
    POSITIVE LOGITS
     parts
    1.02
    もちろん
    0.97
     некоторых
    0.94
     portions
    0.93
     Verw
    0.92
    丁目
    0.91
    끔
    0.90
     توانید
    0.88
    TextUtils
    0.88
     but
    0.88
    Activations Density 0.044%

    No Known Activations