© 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. 9828
    Prev
    Next
    INDEX
    Explanations

    words preceding punctuation

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

    The neuron strongly fires on isolated numeric or single‐character tokens (e.g. years, bracketed numbers, single letters, punctuation)—in short, on odd or “anomalous” standalone tokens rather than ordinary 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
    &:
    0.70
    ệ
    0.66
     cùng
    0.58
    ：
    0.58
    ；
    0.57
     vow
    0.57
    rightsquigarrow
    0.57
    境内
    0.56
    ։
    0.56
    }$.
    0.55
    POSITIVE LOGITS
    HE
    0.73
    HAEL
    0.71
    НЕ
    0.71
    THERE
    0.69
    SOME
    0.67
    чиков
    0.66
     عَل
    0.66
    Ｄ
    0.66
    Ħ
    0.66
    ТЕ
    0.66
    Activations Density 0.009%

    No Known Activations