© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11595
    Prev
    Next
    INDEX
    Explanations

    sanctions against, involved in, extended

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

    The neuron responds strongly to unusually long token fragments (i.e. long subword pieces), effectively flagging very long 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
     गर्ल
    1.00
    데
    0.91
    }}^{-
    0.87
    데요
    0.84
     nonnegative
    0.83
    linson
    0.79
     яе
    0.77
     যৌগ
    0.76
    㳻
    0.76
    ñón
    0.75
    POSITIVE LOGITS
     Sadie
    0.71
     extrac
    0.70
     verific
    0.70
    ни
    0.69
    См
    0.69
     ecce
    0.67
     visant
    0.66
     hyst
    0.66
     grosse
    0.65
     speriment
    0.65
    Activations Density 0.001%

    No Known Activations