© 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. 13428
    Prev
    Next
    INDEX
    Explanations

    exit or leave actions

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

    The neuron activates on words describing someone departing or exiting (e.g. “leaving,” “leaves”).

    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
    电设备
    0.60
     Neighbors
    0.59
    owa
    0.58
     Generations
    0.55
    需求的
    0.54
    genome
    0.54
     extérieures
    0.54
    槅
    0.54
    各个
    0.53
     astronauts
    0.53
    POSITIVE LOGITS
     enters
    0.92
     exit
    0.90
     دخول
    0.88
     enter
    0.85
     entering
    0.83
     entered
    0.82
    進入
    0.80
     confines
    0.79
     exits
    0.78
    enter
    0.77
    Activations Density 0.216%

    No Known Activations