© 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. 10052
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly fires on tokens that represent numeric values (especially decimal numbers).

    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
    б
    0.80
     drop
    0.76
    𝗺
    0.75
     Drop
    0.73
    m
    0.73
     sap
    0.72
     toi
    0.72
     m
    0.72
     sor
    0.70
    b
    0.69
    POSITIVE LOGITS
    IZONTAL
    0.85
     로그인
    0.79
     ordinances
    0.78
    ོང་
    0.78
    Појединачна
    0.77
    (~
    0.77
     liberar
    0.76
     REACTIONS
    0.75
     endeavours
    0.75
     Aztecs
    0.75
    Activations Density 0.000%

    No Known Activations