© 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. 34-GEMMASCOPE-RES-131K
    7. 107320
    Prev
    Next
    INDEX
    Explanations

    Respondent, Include, site

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

    The neuron fires on numeric tokens, spotting sequences of digits (e.g. version numbers, counts, IDs) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     skóry
    -0.74
    yas
    -0.69
    ᡥ
    -0.68
    Ư
    -0.66
     brauchst
    -0.65
    鑫
    -0.64
    XtraGrid
    -0.63
     wish
    -0.62
    닉
    -0.62
    cault
    -0.62
    POSITIVE LOGITS
    ystem
    0.74
    zielle
    0.72
    ūrė
    0.70
     صدا
    0.68
    Bindable
    0.68
    ystems
    0.68
    site
    0.67
    плуата
    0.67
    ристи
    0.66
     Thornton
    0.66
    Activations Density 0.072%

    No Known Activations