© 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. 372
    Prev
    Next
    INDEX
    Explanations

    The neuron is a detector for numeric tokens (i.e. real‐number literals) 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
    어
    0.18
    연
    0.17
    ছিলেন
    0.17
    ق
    0.16
    려고
    0.16
    .(
    0.16
    직
    0.16
    ㅅ
    0.16
    こちらの
    0.16
    angan
    0.16
    POSITIVE LOGITS
     antiquated
    0.23
     aristocracy
    0.19
     필수
    0.19
     unquestion
    0.19
     kaleidoscopic
    0.18
     quintessential
    0.18
     undeniable
    0.18
     unwarranted
    0.18
     compromises
    0.18
     delights
    0.18
    Activations Density 1.090%

    No Known Activations