© 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. 28192
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and variables

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

    The neuron spikes on isolated single‐character tokens (e.g. lone letter variables like “u,” “c,” “j,” etc.).

    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
    Kenapa
    -1.68
    虽说
    -1.62
     När
    -1.61
    Seperti
    -1.56
    Semoga
    -1.52
    Selama
    -1.51
     exuberant
    -1.49
    Meskipun
    -1.47
     meticulously
    -1.46
    Sudah
    -1.46
    POSITIVE LOGITS
    の方も
    1.48
    為に
    1.48
     it
    1.47
    ごとの
    1.44
    際の
    1.34
     ámbitos
    1.29
    の方に
    1.28
    ).
    1.28
    際には
    1.25
    ところです
    1.23
    Activations Density 0.375%

    No Known Activations