© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28037
    Prev
    Next
    INDEX
    Explanations

    say the least, it turns out

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

    The neuron is detecting common function words (e.g. articles, pronouns, auxiliary/modal verbs, and other high-frequency stopwords).

    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
    殃
    -0.97
     to
    -0.91
     like
    -0.91
     €,
    -0.90
     by
    -0.88
     diinginkan
    -0.86
     such
    -0.86
    </h1>
    -0.86
    raktion
    -0.85
     will
    -0.84
    POSITIVE LOGITS
     basteln
    1.18
    brtc
    1.11
    Về
    1.09
     permette
    1.09
    igth
    1.06
     établi
    1.02
     ďal
    1.02
     náš
    1.00
     különböző
    0.99
    Diámetro
    0.99
    Activations Density 0.102%

    No Known Activations