© 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. 39178
    Prev
    Next
    INDEX
    Explanations

    Ker, retry, price, slider, badges, fatal, calendar

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

    The neuron strongly activates on long uppercase identifier-like tokens—acronyms, macro names, constant names, or code symbols.

    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
     fermé
    -1.72
     occupé
    -1.63
     condamné
    -1.52
     accompagné
    -1.50
     arrivée
    -1.50
     isolé
    -1.50
    Zalety
    -1.48
     partiellement
    -1.48
     Kräuter
    -1.46
     Reinigung
    -1.45
    POSITIVE LOGITS
     だっ
    1.82
    -
    1.55
    Interestingly
    1.36
    Also
    1.32
     Their
    1.31
     Их
    1.31
     удалось
    1.30
     After
    1.30
    Other
    1.28
     sonriendo
    1.27
    Activations Density 55.220%

    No Known Activations