© 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. 4913
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and foreign words

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

    The neuron fires strongly on isolated, salient tokens representing domain-specific terms or names (e.g. drug names, product codes, chess notation, etc.).

    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
    י
    1.01
    המ
    0.86
     ain
    0.83
     nuclé
    0.82
    ي
    0.79
     testicule
    0.77
    اعد
    0.77
     comunque
    0.76
     quello
    0.74
    אי
    0.74
    POSITIVE LOGITS
    づくり
    0.83
    ごろ
    0.80
    かり
    0.78
     ночь
    0.73
     должны
    0.72
     атрибу
    0.71
    ྛ
    0.71
     одновременно
    0.70
     способствует
    0.70
    женной
    0.70
    Activations Density 0.000%

    No Known Activations