© 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. 13828
    Prev
    Next
    INDEX
    Explanations

    0, mom, currently, i, global, #, [

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

    The neuron responds most strongly to low‐frequency or unusual tokens—e.g. rare jargon, legal/technical terms, foreign compounds, or otherwise 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
     ему
    0.92
     καθώς
    0.89
    ेंट
    0.86
     κάτι
    0.86
     һәм
    0.85
     contemplation
    0.84
     implementação
    0.83
    ᚑ
    0.83
     Grail
    0.81
     multifaceted
    0.80
    POSITIVE LOGITS
    y
    0.96
    a
    0.93
    yng
    0.92
    o
    0.91
    x
    0.87
    p
    0.84
    sell
    0.83
    iere
    0.82
    no
    0.81
    tor
    0.81
    Activations Density 0.001%

    No Known Activations