© 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. 105210
    Prev
    Next
    INDEX
    Explanations

    IF EXISTS, password

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

    the word "password" in programming and technical contexts.

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

    The neuron fires on tokens that are all uppercase or start with a capital letter—e.g. acronyms, SQL keywords like IF/EXISTS, and proper‐name or acronym tokens.

    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
     that
    -2.30
     to
    -1.84
    
    -1.48
    </h4>
    -1.47
     gydy
    -1.45
     når
    -1.40
    ',
    -1.36
    0
    -1.36
    </h5>
    -1.36
    </td>
    -1.36
    POSITIVE LOGITS
    たまた
    1.41
    atiles
    1.32
    Particular
    1.32
    iyle
    1.31
     combinatie
    1.31
     president
    1.27
    ロマ
    1.27
    olesterol
    1.26
     restaurant
    1.26
    égories
    1.25
    Activations Density 0.074%

    No Known Activations