© 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. 22-GEMMASCOPE-RES-131K
    7. 104524
    Prev
    Next
    INDEX
    Explanations

    Q: or how to

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

    The neuron detects the substring “priv” (as in private or privileged identifiers) within code 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_22/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
     their
    -1.54
     both
    -1.34
    :^{
    -1.16
     mereka
    -1.16
     extenso
    -1.16
     stycznia
    -1.13
    日から
    -1.12
    IORS
    -1.12
     задний
    -1.11
     décembre
    -1.10
    POSITIVE LOGITS
     such
    1.13
    𝑄
    1.09
     installiert
    1.06
    Ingredienti
    1.02
     visitare
    1.01
     остается
    1.01
    нился
    1.01
    Etimología
    1.00
     ilyen
    1.00
     каждому
    0.99
    Activations Density 0.000%

    No Known Activations