© 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. 2748
    Prev
    Next
    INDEX
    Explanations

    resource stress and abuse

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

    The neuron fires on domain-specific technical nouns (e.g. names of treatments, substances, energy/climate/water topics, drugs, carbon, etc.), i.e. specialized jargon.

    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
    側の
    0.51
     splic
    0.50
     facade
    0.50
     spawn
    0.50
    🧊
    0.50
    💠
    0.49
    ここは
    0.49
    🔪
    0.48
     mante
    0.48
     vertebrae
    0.48
    POSITIVE LOGITS
     injustice
    0.66
     disorder
    0.64
     disorders
    0.63
     문제
    0.63
     문제를
    0.63
    问题的
    0.60
     проблемы
    0.58
     abuse
    0.57
     问题
    0.57
     Disorders
    0.56
    Activations Density 0.214%

    No Known Activations