© 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. 36248
    Prev
    Next
    INDEX
    Explanations

    unless otherwise noted or excluded

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

    The neuron detects legal‐boilerplate disclaimer phrases, especially the “without prejudice to … discretion” construction.

    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
     menyen
    -1.26
     hangat
    -1.26
    спасибо
    -1.24
     münchen
    -1.20
     povezave
    -1.18
     背包
    -1.18
     soggior
    -1.17
     diperlukan
    -1.17
     margiela
    -1.15
     收納
    -1.13
    POSITIVE LOGITS
    ,
    1.16
    any
    1.07
     any
    1.06
     natural
    1.06
     how
    1.06
    eno
    1.05
     but
    1.04
     or
    1.04
    Rule
    1.03
    ee
    1.02
    Activations Density 0.008%

    No Known Activations