© 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. 4409
    Prev
    Next
    INDEX
    Explanations

    start with "S"

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

    The neuron fires on rare, technical jargon—especially legal/contract‐style terms (e.g. “Assignment,” “Provisions,” “injunctive,” “processing”)—marking domain-specific vocabulary.

    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
     Treats
    0.95
    Кто
    0.91
    了一个
    0.89
    كي
    0.85
    キ
    0.85
     Mira
    0.84
    𝑫
    0.84
    ва
    0.83
     Amid
    0.83
    アルミ
    0.83
    POSITIVE LOGITS
    خفض
    0.74
    ⎞
    0.73
    ğe
    0.71
    ્ટ
    0.69
    encryption
    0.62
     ujung
    0.61
    ExposureTime
    0.61
     fût
    0.60
    တင်
    0.60
    unregister
    0.60
    Activations Density 0.000%

    No Known Activations