© 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. 10-GEMMASCOPE-RES-131K
    7. 105564
    Prev
    Next
    INDEX
    Explanations

    has been and still is

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

    The neuron fires on occurrences of the copular or auxiliary “to be” verbs (forms like is, are, be, am, was, continues, still, 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_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
     недавно
    -1.32
    并非
    -1.30
    amerikan
    -1.26
     betreffende
    -1.26
     två
    -1.26
     suave
    -1.23
     celebrid
    -1.23
     bidra
    -1.23
     produ
    -1.22
     już
    -1.22
    POSITIVE LOGITS
     Verständnis
    1.59
     usata
    1.33
     könnten
    1.33
     anderes
    1.28
     During
    1.27
     teilweise
    1.23
    之心
    1.23
    sistors
    1.23
    之子
    1.20
     Aussehen
    1.18
    Activations Density 0.022%

    No Known Activations