© 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. 25647
    Prev
    Next
    INDEX
    Explanations

    closing remarks and names

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

    The neuron fires on email sign-off lines, footers, and separators (e.g. “thanks,” “regards,” “----- Original Message -----”).

    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
     singkat
    -1.16
     anaknya
    -1.12
     pertemuan
    -1.09
    葵
    -1.06
     KNOW
    -1.04
     seluruh
    -1.01
     komplett
    -1.00
     kesä
    -0.99
    område
    -0.99
     tambah
    -0.99
    POSITIVE LOGITS
    p
    1.09
    ↵
    1.05
    ↵↵
    1.03
    :
    1.02
     demselben
    0.98
    了
    0.97
    !
    0.95
    农村
    0.94
    就可以了
    0.92
     divertida
    0.91
    Activations Density 0.291%

    No Known Activations