© 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. 13526
    Prev
    Next
    INDEX
    Explanations

    file paths and code structure

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

    The neuron fires strongly on tokens containing digits (e.g. numeric IDs, version numbers, file sizes), i.e. it detects numbers embedded in tokens.

    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
    embro
    -1.18
    levis
    -1.16
     can
    -1.13
    toner
    -1.05
    perry
    -1.05
    Nhap
    -1.04
    cilia
    -1.01
    }`,
    -1.01
     cromado
    -1.00
     ενώ
    -0.99
    POSITIVE LOGITS
    _
    1.33
    場合には
    1.30
    你不是
    1.22
     pantalon
    1.20
    𝑵
    1.13
     dîner
    1.13
    Ƭ
    1.12
    entsch
    1.09
    برای
    1.09
     snart
    1.09
    Activations Density 0.061%

    No Known Activations