© 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. 10-GEMMASCOPE-RES-131K
    7. 15434
    Prev
    Next
    INDEX
    Explanations

    erroneous or bogus information

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

    The neuron flags occurrences of “erroneous” (and its adverb form “erroneously”).

    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
     say
    -2.03
    -1.88
    雷霆
    -1.88
                
    -1.86
    (
    -1.84
    So
    -1.84
     cuando
    -1.83
    可以
    -1.81
     Honorable
    -1.81
     WAY
    -1.80
    POSITIVE LOGITS
    淠
    2.52
    
    2.28
    が大
    2.28
    特意
    2.25
     間取り
    2.20
    玙
    2.16
    ～～～
    2.14
    ーーーー
    2.14
    》：
    2.14
    ～　
    2.13
    Activations Density 0.003%
    No Known Activations