© 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. 9324
    Prev
    Next
    INDEX
    Explanations

    column definition

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

    The neuron responds most strongly to unusually rare or domain‐specific tokens (i.e. those with very high TF–IDF weight).

    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
    ována
    0.80
    傥
    0.77
     cập
    0.71
     lau
    0.71
     നിന്നും
    0.70
     Ausnahme
    0.70
     recycl
    0.70
    issão
    0.70
     newState
    0.70
     bloqueo
    0.70
    POSITIVE LOGITS
     weddings
    0.88
     championships
    0.82
     sweetly
    0.82
     veggies
    0.80
     bullies
    0.79
    e
    0.79
    Е
    0.78
    €™
    0.77
     wrists
    0.77
     qualifiers
    0.76
    Activations Density 0.001%

    No Known Activations