© 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. 73753
    Prev
    Next
    INDEX
    Explanations

    titles and questions

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

    The neuron specializes in spotting ASCII-art style diagrams or block diagrams made up of dashes, pipes, letters, and numbers.

    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
     for
    -1.68
     our
    -1.30
     of
    -1.28
     that
    -1.27
     may
    -1.23
     as
    -1.22
     because
    -1.16
     Many
    -1.16
     often
    -1.13
     those
    -1.12
    POSITIVE LOGITS
     sembler
    1.29
     Cruelty
    1.16
    粜
    1.15
     bénéficier
    1.13
    ₤
    1.10
     apparaît
    1.09
     maniere
    1.09
     actuellement
    1.09
    itati
    1.06
     Skilled
    1.06
    Activations Density 0.003%

    No Known Activations