© 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. 10-GEMMASCOPE-RES-131K
    7. 59268
    Prev
    Next
    INDEX
    Explanations

    Q, The, Introduction, Legal, Court

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

    The neuron consistently activates on tokens containing uppercase letters—proper names, technology/library identifiers, or other capitalized words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    猡
    -1.73
    Taman
    -1.50
    Desarrollo
    -1.48
     disiplin
    -1.45
    朥
    -1.44
    metodo
    -1.40
    sandalia
    -1.38
    Frauen
    -1.38
    fichier
    -1.38
    蕯
    -1.36
    POSITIVE LOGITS
    [
    1.77
    =
    1.70
    0
    1.50
    するのが
    1.49
    ]
    1.48
    <em>
    1.46
    <strong>
    1.46
    Mark
    1.44
     tutaj
    1.44
    ここで
    1.43
    Activations Density 0.002%

    No Known Activations