© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 79
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively fires on tokens representing floating-point or other multi-digit 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
    verständlich
    0.31
    ました
    0.31
    <unused973>
    0.30
    лся
    0.29
    ろ
    0.28
    ровку
    0.28
    钎
    0.26
     ло
    0.26
    อย่าง
    0.26
    어요
    0.26
    POSITIVE LOGITS
    ia
    0.37
    -
    0.36
    z
    0.36
    al
    0.35
    an
    0.34
    el
    0.32
    ul
    0.32
    ०
    0.31
    un
    0.31
    ch
    0.30
    Activations Density 0.000%

    No Known Activations