© 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. 12190
    Prev
    Next
    INDEX
    Explanations

    parentheses followed by comma

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

    The neuron detects hidden Unicode bidirectional‐text control characters (e.g. the LTR/RTL override marks).

    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
    AT
    1.23
    ED
    1.20
     blaze
    1.12
    在
    1.10
    ak
    1.09
    ET
    1.06
     বললেন
    1.03
     blazing
    1.03
    ين
    1.00
    شون
    1.00
    POSITIVE LOGITS
    ات
    1.64
    ت
    1.34
    s
    1.30
    ्स
    1.23
    tır
    1.20
    ્સ
    1.15
    ς
    1.12
    cita
    1.11
     Einstellungen
    1.07
    я
    1.07
    Activations Density 0.445%

    No Known Activations