© 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. 14247
    Prev
    Next
    INDEX
    Explanations

    This neuron responds to unusually rare or low-frequency tokens (i.e. high-IDF or domain-specific words and symbols).

    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
    而且
    0.77
    。「
    0.69
    amaran
    0.69
    并且
    0.67
    並且
    0.66
    。
    0.64
    %。
    0.63
    ។
    0.63
     siebie
    0.62
    บ้าง
    0.62
    POSITIVE LOGITS
     its
    1.20
     Its
    1.06
     associated
    1.05
    its
    1.00
    Its
    0.98
     related
    0.91
    associated
    0.82
     அதன்
    0.79
     উহার
    0.77
     corresponding
    0.75
    Activations Density 0.677%

    No Known Activations