© 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. 10223
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently lights up on numeric tokens—especially floating-point numbers—such as upload times, view counts, or other metric 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
    lden
    0.80
     LCD
    0.74
    y
    0.74
    란드
    0.73
    ia
    0.71
    ardi
    0.71
    ampe
    0.70
    nbsp
    0.69
     candidate
    0.68
    uin
    0.67
    POSITIVE LOGITS
    พ
    0.91
    हमारी
    0.85
    Steph
    0.81
    ภ
    0.81
    ศึกษา
    0.80
    淒
    0.80
    เซ
    0.79
    Imagery
    0.78
    Chúng
    0.78
    ธ
    0.78
    Activations Density 0.001%

    No Known Activations