© 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. 34-GEMMASCOPE-RES-131K
    7. 8307
    Prev
    Next
    INDEX
    Explanations

    frequently asked questions

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

    The neuron fires on the words and phrases used in FAQ headings—e.g. “frequently,” “asked,” “questions,” “answered” and “list”—i.e. it detects “Frequently Asked Questions” style labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     inkább
    -0.70
     SIGN
    -0.69
     đội
    -0.68
    signing
    -0.68
     signatures
    -0.67
     steers
    -0.67
     Signature
    -0.66
    wój
    -0.66
    чним
    -0.66
    Адрес
    -0.65
    POSITIVE LOGITS
     brim
    0.69
    元に
    0.67
     pflegen
    0.65
     गल
    0.63
    ัน
    0.63
    roof
    0.63
    Pb
    0.63
    })();
    0.62
     德
    0.61
     Mond
    0.61
    Activations Density 0.062%

    No Known Activations