© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103874
    Prev
    Next
    INDEX
    Explanations

    citation markers like [@B1]

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

    This neuron activates on numeric reference markers and numbers in parentheses or brackets (e.g., citation indices and parenthetical numeric values).

    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
     your
    -1.56
     until
    -1.41
     llega
    -1.40
     felpa
    -1.38
     delicately
    -1.37
    为啥
    -1.35
     subtle
    -1.34
    平常
    -1.33
     urging
    -1.30
     needed
    -1.29
    POSITIVE LOGITS
     Juillet
    1.48
     but
    1.43
    .
    1.41
    읗
    1.34
     Octobre
    1.33
    toso
    1.33
     karier
    1.32
     DECEMBER
    1.32
    брав
    1.32
    เที่ยว
    1.30
    Activations Density 0.041%

    No Known Activations