© 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. 11935
    Prev
    Next
    INDEX
    Explanations

    higher, larger, per

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

    The neuron fires on unusually long tokens—i.e. multi‐syllable or high‐character‐count 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
    ivating
    0.88
    ussians
    0.88
    leting
    0.86
    ționale
    0.86
    iguing
    0.84
    푐
    0.84
    ੌਰ
    0.83
    vamos
    0.83
     weren
    0.82
    ម្បី
    0.81
    POSITIVE LOGITS
     Kamu
    0.80
    Ś
    0.75
     Aula
    0.74
    Kamu
    0.73
     Mme
    0.72
    Chiesa
    0.71
    Aula
    0.70
     Jews
    0.70
     dû
    0.69
    LCD
    0.68
    Activations Density 0.000%

    No Known Activations