© 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. 4757
    Prev
    Next
    INDEX
    Explanations

    The neuron chiefly responds to rare or specialized tokens—uncommon words, acronyms, proper names, and other low‐frequency multi‐syllable terms.

    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
    _
    0.96
    it
    0.89
    enable
    0.81
    ;
    0.79
    },
    0.79
    ordinary
    0.78
    generic
    0.76
    leri
    0.75
    有的
    0.75
    include
    0.74
    POSITIVE LOGITS
    ᒪ
    0.91
     вопроса
    0.90
     antérieurs
    0.89
     Einkauf
    0.84
     Эта
    0.82
     जेईई
    0.81
    閇
    0.81
    Ｕ
    0.81
     ウォ
    0.80
    ᒻ
    0.80
    Activations Density 0.000%

    No Known Activations