© 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. 5973
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on words that signal withheld or confidential details—e.g. anonymity, private, later—i.e. markers of information being kept undisclosed or deferred.

    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
     muffled
    0.55
     impressive
    0.53
     unbeaten
    0.47
     ఓ
    0.47
     lose
    0.46
     mensaje
    0.46
     ingenious
    0.46
     одной
    0.46
     tale
    0.45
    выше
    0.45
    POSITIVE LOGITS
    เพื่อ
    0.60
    。《
    0.57
    。“
    0.55
    ۔
    0.55
    ೮
    0.55
     because
    0.54
    ជាមួយនឹង
    0.54
     क्योंकि
    0.53
    ().
    0.52
    针对
    0.51
    Activations Density 0.000%

    No Known Activations