© 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. 12122
    Prev
    Next
    INDEX
    Explanations

    sentence end punctuation

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

    The neuron strongly activates on proper‐name tokens (especially names and organizations) and on punctuation marks that signal major clause or sentence boundaries.

    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
     слід
    0.93
    Namun
    0.89
     НЕ
    0.85
     лише
    0.84
     alcançar
    0.81
     หาก
    0.81
    Após
    0.79
     azonban
    0.77
    Однако
    0.77
     вже
    0.77
    POSITIVE LOGITS
     그다음에
    1.19
     그러니까
    1.05
    っていう
    0.89
    基本上
    0.88
     굉장히
    0.87
     [
    0.84
     yeah
    0.82
     veldig
    0.82
     basically
    0.82
     Yeah
    0.79
    Activations Density 0.335%

    No Known Activations