© 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. 31-GEMMASCOPE-2-RES-16K
    7. 13548
    Prev
    Next
    INDEX
    Explanations

    slow or slower speeds

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

    The neuron strongly activates on occurrences of the word “slow.”

    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
    ရပ်
    0.61
     quickly
    0.60
    ible
    0.58
     leap
    0.58
     Shortcuts
    0.58
     Quickly
    0.57
     뛰
    0.57
     leaped
    0.57
    快捷
    0.56
    shortcuts
    0.55
    POSITIVE LOGITS
     slow
    2.81
     Slow
    2.64
    Slow
    2.61
    slow
    2.55
     slower
    2.34
     slowest
    2.30
    慢
    2.23
    SLOW
    2.19
     slowed
    2.14
    缓慢
    2.14
    Activations Density 0.279%

    No Known Activations