© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 63550
    Prev
    Next
    INDEX
    Explanations

    categories and preparation

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

    The neuron strongly fires on tokens that appear in headlines or news-wire reports about disease outbreaks and treatments—especially Ebola-related case and patient terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ℕ
    -0.82
    ſo
    -0.72
    Klein
    -0.72
    Workplace
    -0.70
    sword
    -0.69
    kiat
    -0.69
    ßten
    -0.68
     Sediment
    -0.68
    ὄ
    -0.68
     ПА
    -0.68
    POSITIVE LOGITS
     suit
    1.21
     suits
    1.14
     Level
    1.04
     haz
    0.94
     lvl
    0.87
     Suit
    0.87
    Level
    0.86
     Suits
    0.85
    enguin
    0.85
     Haz
    0.85
    Activations Density 0.021%

    No Known Activations