© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13879
    Prev
    Next
    INDEX
    Explanations

    emergency services

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

    The neuron strongly activates on uppercase initialisms or acronyms (e.g. “AED,” “CPR,” “ACL,” “Sgt,” etc.).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     бухгалтер
    0.63
    襖
    0.62
     pitted
    0.58
    buri
    0.58
    审计
    0.57
     noncommutative
    0.57
     Ibar
    0.55
    ardust
    0.54
     Nagasaki
    0.54
     Boutique
    0.54
    POSITIVE LOGITS
     emergency
    1.80
     Emergency
    1.80
    Emergency
    1.78
    救
    1.71
    emergency
    1.70
     rescuers
    1.70
     ambulance
    1.70
     rescue
    1.66
     EMERGENCY
    1.61
     Ambulance
    1.59
    Activations Density 0.331%

    No Known Activations