© 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. 9632
    Prev
    Next
    INDEX
    Explanations

    list items with structure

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

    The neuron detects object property names or keys (identifiers immediately followed by a colon) in code or JSON‐like structures.

    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
     also
    -1.25
     モード
    -0.99
     також
    -0.99
    Leírás
    -0.95
     would
    -0.92
    exprimer
    -0.91
    Preparazione
    -0.91
     arī
    -0.91
    Caracteristici
    -0.91
    Materiale
    -0.90
    POSITIVE LOGITS
     первая
    1.20
     eerste
    1.20
     pierws
    1.08
     ensimmä
    1.08
     första
    1.01
     primeros
    0.99
     timmar
    0.98
     первого
    0.98
     дори
    0.97
     first
    0.96
    Activations Density 0.062%

    No Known Activations