© 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. 34-GEMMASCOPE-RES-131K
    7. 97911
    Prev
    Next
    INDEX
    Explanations

    security incidents and attacks

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

    The neuron fires strongly on tokens that look like section headings, metadata, or other prominent labels—e.g. uppercase words, acronyms, dates, numbers, and key technical 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
    𝓪
    -0.75
     seguras
    -0.74
    Estamos
    -0.74
     religieuse
    -0.72
     Apakah
    -0.71
    倉
    -0.71
    듦
    -0.71
    必须要
    -0.69
     özg
    -0.69
     projecting
    -0.68
    POSITIVE LOGITS
     SIEM
    1.42
     incident
    1.33
     responder
    1.14
     correlation
    1.12
     investigations
    1.11
     investigation
    1.09
     hunt
    1.09
     detections
    1.09
     Incident
    1.06
     Investigations
    1.06
    Activations Density 0.034%

    No Known Activations