© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14962
    Prev
    Next
    INDEX
    Explanations

    steps and code structure

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

    The neuron fires most strongly on relatively rare or unusual tokens (e.g. low‐frequency words or standalone characters) in the text.

    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
    atthena
    0.88
    dollars
    0.87
     vitth
    0.86
    ovaniyu
    0.86
     elektronik
    0.86
    𝓸
    0.86
     dukkham
    0.84
    ொருள்
    0.84
     etkinlik
    0.84
    nesday
    0.82
    POSITIVE LOGITS
    ’
    0.85
    ع
    0.72
     ২৬
    0.67
    phus
    0.66
    ama
    0.66
     ৩১
    0.63
     lab
    0.63
     cri
    0.62
     stair
    0.62
     สุด
    0.61
    Activations Density 0.000%

    No Known Activations