© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 87210
    Prev
    Next
    INDEX
    Explanations

    bombing, shooting, fire, or explosion events

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

    The neuron strongly activates on words naming violent or explosive incidents (e.g. “explosion,” “blast,” “shooting,” “bomb,” “fire,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    (!__
    -1.70
    xically
    -1.68
     و
    -1.55
    '}}>
    -1.55
    ivable
    -1.55
    व
    -1.53
    最初
    -1.52
    ون
    -1.49
     лише
    -1.49
     וב
    -1.44
    POSITIVE LOGITS
     in
    2.02
    シャレ
    1.65
     sorprende
    1.53
    ↵↵
    1.52
    さまざま
    1.52
     habia
    1.51
    有两
    1.49
     separates
    1.48
    jadikan
    1.46
    そも
    1.45
    Activations Density 0.067%

    No Known Activations