© 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. 26843
    Prev
    Next
    INDEX
    Explanations

    administration, policy, safety, travel

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

    The neuron responds to formal news‐reporting language—tokens you’d see when an official decision or statement is being described (e.g. “decision,” “reason,” “administration,” “says,” numbers/percentages).

    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
     lilla
    -1.05
    IDENTAL
    -1.04
    '))
    -1.03
    Ἅ
    -1.03
    ♡♡
    -1.02
     forêts
    -1.01
     ı
    -0.99
     gelas
    -0.99
    `
    -0.98
    pæ
    -0.98
    POSITIVE LOGITS
     safety
    1.33
     safe
    1.23
    ּוֹ
    1.13
     our
    1.12
     Safety
    1.12
     PRODUCTO
    1.09
     încă
    1.06
     INFORMACIÓN
    1.06
     travel
    1.06
     August
    1.06
    Activations Density 0.012%

    No Known Activations