© 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. 7009
    Prev
    Next
    INDEX
    Explanations

    lateral scaling, specialization, swap, coups, regex, boost

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

    The neuron strongly activates on uncommon or low‐frequency tokens (i.e. rare words or subwords).

    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
     advogado
    0.71
     "'
    0.70
     COPYRIGHT
    0.67
     Fakat
    0.66
    0.66
     década
    0.66
     noticias
    0.66
     \{\
    0.66
     ["
    0.64
    news
    0.64
    POSITIVE LOGITS
     permangan
    0.86
     atro
    0.85
     sprinkler
    0.85
     cloison
    0.83
     Peloton
    0.83
     slipper
    0.81
     plumage
    0.80
    ЛЬ
    0.79
     zaten
    0.79
     sprink
    0.79
    Activations Density 0.000%

    No Known Activations