© 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. 12875
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on tokens that have very high vocabulary index IDs (i.e. rare or high-numbered byte-pair encoding tokens).

    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
    뻑
    0.90
     गाड़ियों
    0.87
    太过
    0.84
     sgt
    0.83
     perfectly
    0.83
     parfaitement
    0.83
     macho
    0.82
     cuarto
    0.81
     motorist
    0.81
     superfluous
    0.81
    POSITIVE LOGITS
    8
    0.76
     А
    0.73
    7
    0.73
    9
    0.69
    6
    0.69
     ADS
    0.68
    şil
    0.66
    Biod
    0.65
     Medical
    0.65
    ุ
    0.65
    Activations Density 0.001%

    No Known Activations