© 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. 87077
    Prev
    Next
    INDEX
    Explanations

    redux import storage id

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

    This neuron strongly activates on numeric tokens and underscores in code—especially version numbers embedded in package or import paths.

    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
     consé
    -2.52
     viſ
    -2.47
    が
    -2.42
    ᜀ
    -2.41
    ～！
    -2.41
    を
    -2.41
     花朵
    -2.38
     炖
    -2.36
     linge
    -2.31
     fré
    -2.28
    POSITIVE LOGITS
    M
    2.80
    N
    2.67
    T
    2.67
    F
    2.64
    O
    2.64
    L
    2.63
    b
    2.61
    t
    2.58
    U
    2.56
    為に
    2.53
    Activations Density 55.432%

    No Known Activations