© 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. 104350
    Prev
    Next
    INDEX
    Explanations

    AES encryption standard

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

    the name "Diego" or references to AES encryption/cryptography.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron strongly activates on tokens containing internal capitalization—e.g. acronyms and CamelCase or PascalCase identifiers.

    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
    袮
    -3.59
    u
    -2.72
     glyphicon
    -2.67
    ”
    -2.64
    酲
    -2.42
    悒
    -2.36
    𓐍
    -2.34
    rierten
    -2.33
     शुर
    -2.31
    餮
    -2.31
    POSITIVE LOGITS
     It
    3.06
    <bos>
    2.88
    {
    2.73
    
    2.30
     le
    2.30
    -
    2.25
     But
    2.22
     Then
    2.16
     Таким
    2.16
     waren
    2.14
    Activations Density 0.004%

    No Known Activations