© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13301
    Prev
    Next
    INDEX
    Explanations

    language constructs like `where` or `class`

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

    The neuron detects numeric tokens or figures (e.g. years, measurements, prices) 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
    ଓ
    0.86
    ୪
    0.82
    筫
    0.81
    され
    0.80
    چی
    0.80
    匏
    0.80
    贺
    0.78
    さらに
    0.77
    Һ
    0.77
    英国
    0.76
    POSITIVE LOGITS
    t
    0.98
    es
    0.92
    ecc
    0.88
     Redemption
    0.88
    s
    0.88
    elevation
    0.82
    isomer
    0.81
    h
    0.79
    is
    0.79
    sulph
    0.78
    Activations Density 0.001%

    No Known Activations