© 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. 10484
    Prev
    Next
    INDEX
    Explanations

    complexity and scale of things

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

    The neuron strongly activates on numeric expressions—i.e. tokens representing numbers or numeric 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
     k
    1.09
     I
    1.06
     /
    1.01
     K
    0.99
     m
    0.98
     D
    0.96
     A
    0.92
     .
    0.90
     s
    0.89
     h
    0.89
    POSITIVE LOGITS
     দেশটির
    0.95
     estadounidenses
    0.91
     بسبب
    0.88
     সেখানকার
    0.88
    它们
    0.87
    äischen
    0.86
    被称为
    0.86
     चीजों
    0.85
    数百
    0.84
    他们的
    0.83
    Activations Density 0.912%

    No Known Activations