© 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. 22-GEMMASCOPE-RES-131K
    7. 35324
    Prev
    Next
    INDEX
    Explanations

    metadata keys

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

    The neuron fires on JSON (or JS object) property names—i.e. the keys like "name", "version", "license", "private", etc.

    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
     semelh
    -1.16
     VIDEOTAPE
    -1.15
     treino
    -1.09
     своем
    -1.08
    €™
    -1.07
    ʻ
    -1.06
    带领
    -1.05
     consentimento
    -1.05
     Ähn
    -1.03
     castelo
    -1.03
    POSITIVE LOGITS
     in
    1.14
    Lá
    1.03
    Jangan
    1.02
     усе
    1.00
    낀
    0.99
     melindungi
    0.99
    4
    0.96
    8
    0.96
    ！！」
    0.96
     Saxe
    0.95
    Activations Density 0.069%

    No Known Activations