© 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. 16179
    Prev
    Next
    INDEX
    Explanations

    The neuron flags unusually rare or infrequently seen tokens in the text, essentially highlighting low‐frequency (uncommon) words.

    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
    ا
    1.15
     Mosul
    0.86
    ջ
    0.80
     tahmin
    0.76
    秚
    0.76
    ї
    0.75
    الہ
    0.75
     Antioch
    0.75
    >∕
    0.75
     clientèle
    0.74
    POSITIVE LOGITS
    ART
    0.88
    toc
    0.87
    tions
    0.85
    sg
    0.81
    t
    0.80
    taro
    0.79
    Ans
    0.78
    tional
    0.76
    the
    0.75
    OR
    0.75
    Activations Density 0.000%

    No Known Activations