© 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. 10-GEMMASCOPE-RES-131K
    7. 104446
    Prev
    Next
    INDEX
    Explanations

    Refer

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

    The neuron strongly activates on numeric literals (especially multi‐digit numbers) in code or data listings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ৢ
    -1.84
    *}[
    -1.70
    xyl
    -1.69
    xvii
    -1.66
    vaného
    -1.63
    \_
    -1.58
    ichteten
    -1.58
    aruhi
    -1.56
     &
    -1.56
    ichtete
    -1.55
    POSITIVE LOGITS
     also
    1.55
     bästa
    1.50
    Varian
    1.45
    Klem
    1.43
     såg
    1.43
    Variante
    1.42
     ktoré
    1.41
    Upvote
    1.38
    Whisper
    1.38
     förstå
    1.38
    Activations Density 0.012%

    No Known Activations