© 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. 8954
    Prev
    Next
    INDEX
    Explanations

    AI, algorithm, improvement

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

    The neuron responds strongly to unusually rare or out-of-vocabulary tokens (e.g. numbers, proper names, odd punctuation), effectively flagging low-frequency or “weird” 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
    ’
    0.85
    </
    0.84
    ia
    0.84
    ity
    0.83
    ard
    0.80
    can
    0.80
    yn
    0.79
    is
    0.77
    son
    0.76
    ical
    0.75
    POSITIVE LOGITS
     funzionalità
    0.95
    앱
    0.85
    经费
    0.82
     pouvoirs
    0.81
    ప్ర
    0.81
     longitudinale
    0.80
     biopsies
    0.79
     класу
    0.79
     quinze
    0.79
    랴
    0.79
    Activations Density 0.000%

    No Known Activations