© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128607
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens (especially version numbers or other explicit numeric values).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    electrokinetic remediation

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

    markup or code syntax in technical documents and configuration files.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     it
    -1.48
     this
    -1.38
     both
    -1.28
     from
    -1.23
     their
    -1.16
     does
    -1.16
     its
    -1.16
     several
    -1.11
     on
    -1.10
     by
    -1.10
    POSITIVE LOGITS
    abetes
    1.09
     diſ
    1.08
    这些
    1.05
     exemplu
    1.03
    these
    1.02
    нес
    1.01
     urat
    1.00
     Эти
    1.00
    Metallica
    0.99
    這些
    0.98
    Activations Density 0.000%

    No Known Activations