© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5544
    Prev
    Next
    INDEX
    Explanations

    Parse.Object.extend("Car")

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

    The neuron fires on uppercase command‐style tokens (macro or API calls and keywords) in the text.

    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
     mois
    0.64
     사는
    0.63
     Wales
    0.61
     tien
    0.61
     tổng
    0.61
     we
    0.60
     항
    0.59
     (?)
    0.59
     따른
    0.59
     had
    0.58
    POSITIVE LOGITS
    あえず
    0.76
    create
    0.70
    initialize
    0.68
    Group
    0.64
    ('$
    0.64
    mutable
    0.64
    ValueSet
    0.64
    Property
    0.64
    さらに
    0.61
    
    0.60
    Activations Density 0.043%

    No Known Activations