© 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. 36400
    Prev
    Next
    INDEX
    Explanations

    numbers followed by entities or units

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

    The neuron activates on numeric quantity expressions, i.e. numbers together with their unit or category (people, countries, million, years, miles, cm, etc.).

    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
     as
    -1.27
     only
    -1.21
    と思いました
    -1.18
    块钱
    -1.18
     etc
    -1.18
    fourths
    -1.14
     most
    -1.11
     several
    -1.09
     restos
    -1.09
     provide
    -1.07
    POSITIVE LOGITS
    場合には
    1.30
    の方に
    1.28
    ことも
    1.28
    明日は
    1.20
    さぁ
    1.16
    定価
    1.15
    が欲しい
    1.12
    はこんな感じ
    1.11
     tenderly
    1.10
    さっそく
    1.10
    Activations Density 0.072%

    No Known Activations