© 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. 4729
    Prev
    Next
    INDEX
    Explanations

    modifiable, learnable, applicable

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

    This neuron flags anomalous or out‐of‐distribution tokens—especially stray dashes or otherwise unusually formatted/rare tokens—that stand out from normal 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
    čkom
    0.83
     risk
    0.82
     Haber
    0.80
    る
    0.79
    ،
    0.79
     counterclaim
    0.77
     gefallen
    0.77
    Viva
    0.76
     ڈپاز
    0.75
     timeframe
    0.75
    POSITIVE LOGITS
    ּ
    0.78
     полови
    0.75
     aussitôt
    0.73
    cil
    0.71
     itertools
    0.70
     सहयोगी
    0.70
     आर
    0.70
     difíc
    0.68
    嗓
    0.68
     đôi
    0.66
    Activations Density 0.000%

    No Known Activations