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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65825
    Prev
    Next
    INDEX
    Explanations

    phrases starting with word types

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

    The neuron detects passages describing security/privacy measures or handling of sensitive customer information.

    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
     teka
    -1.30
     acidente
    -1.18
     Uhr
    -1.08
     perigo
    -1.07
     povo
    -1.06
     espanhol
    -1.06
     kesi
    -1.05
     namorado
    -1.04
     respondió
    -1.02
     ucap
    -1.01
    POSITIVE LOGITS
    意義
    1.13
    됐
    1.12
     naše
    1.11
    おはよう
    1.08
    これも
    1.07
     supposing
    1.07
    čnosti
    1.07
     možnost
    1.06
    あなたが
    1.06
     numeral
    1.05
    Activations Density 0.047%

    No Known Activations