© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14102
    Prev
    Next
    INDEX
    Explanations

    programming keywords and symbols

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

    The neuron detects numeric tokens or sequences (e.g., version numbers, hex color codes, dimensions, and other digit strings).

    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
    m
    -1.24
    s
    -1.20
     einen
    -1.20
     cette
    -1.13
    I
    -1.12
    p
    -1.12
     samego
    -1.04
    n
    -1.03
    al
    -1.01
    adari
    -1.01
    POSITIVE LOGITS
     and
    1.73
    Baru
    1.30
    Suka
    1.28
    Teknik
    1.21
    samt
    1.15
    ebenarnya
    1.13
     آمریک
    1.13
    Fakta
    1.11
    Baik
    1.10
    Legende
    1.09
    Activations Density 0.172%

    No Known Activations