© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 80050
    Prev
    Next
    INDEX
    Explanations

    numbers and equations

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

    The neuron is primarily detecting sequences of digits (numeric tokens) in the text.

    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
     majest
    -1.24
     capot
    -1.18
    Inggris
    -1.18
     costura
    -1.16
     vernac
    -1.15
     AtomicInteger
    -1.13
    >−</
    -1.12
     když
    -1.10
     obed
    -1.09
     AÑO
    -1.09
    POSITIVE LOGITS
     also
    1.55
    M
    1.30
    4
    1.26
    8
    1.20
    6
    1.19
    R
    1.16
    u
    1.13
    H
    1.13
    N
    1.10
    9
    1.10
    Activations Density 0.093%

    No Known Activations