© 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. 75404
    Prev
    Next
    INDEX
    Explanations

    start of introduction

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

    This neuron fires on lines made up of repeated “=” characters that serve as section dividers or heading underlines.

    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
    葙
    -2.70
    咵
    -2.44
     sprüche
    -2.27
    蒾
    -2.14
     malheur
    -2.13
    
    -2.13
    ܜ
    -2.11
    '!
    -2.03
     bois
    -2.03
    😫
    -2.00
    POSITIVE LOGITS
    0
    3.48
     —
    2.64
    2
    2.58
    箇所
    2.56
    where
    2.48
    which
    2.27
    no
    2.25
    4
    2.23
     auß
    2.20
    n
    2.19
    Activations Density 0.005%

    No Known Activations