© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13723
    Prev
    Next
    INDEX
    Explanations

    context punctuation or digits

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

    This neuron detects numeric tokens representing decimal (floating-point) values.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     *"
    0.88
    য়ী
    0.86
     ủy
    0.84
     Marketplace
    0.80
     DataLoader
    0.79
     adversity
    0.77
     WebSocket
    0.76
    يفي
    0.76
     Borussia
    0.75
    
    0.75
    POSITIVE LOGITS
    Longueur
    0.72
     পুক
    0.67
    โฟ
    0.66
    FileURL
    0.66
    brellas
    0.65
    ্ড
    0.65
     matrícula
    0.65
    acc
    0.64
    bestos
    0.64
     sheep
    0.64
    Activations Density 0.001%

    No Known Activations