© 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. 31-GEMMASCOPE-2-RES-16K
    7. 9410
    Prev
    Next
    INDEX
    Explanations

    often, frequently, or repeatedly

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

    The neuron strongly activates on sentence-boundary cues—especially end-of-sentence punctuation (periods) and the surrounding capitalization or punctuation marks.

    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_31_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
    roadmap
    0.68
    studies
    0.66
    書籍
    0.66
    funding
    0.64
    currently
    0.64
    Books
    0.63
    授
    0.62
    books
    0.62
    improvements
    0.61
    🎤
    0.61
    POSITIVE LOGITS
     চমৎকার
    0.72
     Delicate
    0.71
     delicate
    0.71
     elementów
    0.66
     délic
    0.66
     exquisitely
    0.65
     superbly
    0.64
     Rarely
    0.64
     Excellent
    0.64
     Dalib
    0.64
    Activations Density 0.189%

    No Known Activations