© 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. 22-GEMMASCOPE-RES-131K
    7. 98177
    Prev
    Next
    INDEX
    Explanations

    that...meatthey...onlythat...shedefendant...’Jedi...engagedthat...suchthat...shewas into havePitch ingBacter ol

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

    This neuron detects the legal-style evidentiary language used to introduce proof requirements—especially subordinating markers like “that” and verbs such as “prove,” “show,” or “establish” in a legal context.

    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
     to
    -1.05
     voudrais
    -1.02
     schien
    -0.96
     chrétienne
    -0.94
    árna
    -0.91
     排行
    -0.90
     utafiti
    -0.89
     perpetrators
    -0.88
     Yahoo
    -0.88
     فقط
    -0.88
    POSITIVE LOGITS
     sufficient
    1.27
     how
    1.21
     adequate
    1.18
     appropriate
    1.03
     enough
    1.00
     why
    0.98
     attempted
    0.98
     their
    0.98
     reasonable
    0.96
     used
    0.95
    Activations Density 0.017%

    No Known Activations