© 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. 34034
    Prev
    Next
    INDEX
    Explanations

    code block markers

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

    The neuron is triggered by programming identifiers (variable and function names) rather than keywords, punctuation, or ordinary English words.

    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
     Hitze
    -0.88
     コミック
    -0.88
    TAINER
    -0.86
    сюда
    -0.81
    летний
    -0.80
    tiglia
    -0.79
     ۳۰
    -0.79
     aix
    -0.79
     sebe
    -0.78
    कार
    -0.78
    POSITIVE LOGITS
     semplicemente
    0.91
     decât
    0.90
    Добавлено
    0.88
     esattamente
    0.85
     SAA
    0.84
    День
    0.84
     nemmeno
    0.82
    Хочу
    0.81
     wiederum
    0.81
    有一
    0.81
    Activations Density 0.018%

    No Known Activations