© 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. 34-GEMMASCOPE-RES-131K
    7. 58347
    Prev
    Next
    INDEX
    Explanations

    closing and sign-off phrases

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

    This neuron detects document‐level metadata and formatting tokens—especially numbers (dates, times, “8th,” “25.11.2019,” etc.), punctuation marks, and other structural or signature‐style elements.

    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
    Buongiorno
    -0.87
     Superhosts
    -0.85
    ikon
    -0.81
     delante
    -0.81
    fancybox
    -0.80
    ongiorno
    -0.80
     Celebration
    -0.80
    sile
    -0.78
     anhand
    -0.78
    Werde
    -0.77
    POSITIVE LOGITS
    Sincerely
    3.69
     Sincerely
    3.53
    Yours
    2.77
     sincerely
    2.61
     Yours
    2.56
     yours
    2.47
     Regards
    2.36
    Regards
    2.23
    Signed
    2.14
     Respectfully
    2.09
    Activations Density 0.029%

    No Known Activations