© 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. 35572
    Prev
    Next
    INDEX
    Explanations

    incorrectness, misunderstanding, or mistakes

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

    The neuron responds to words signaling mistakes or misunderstanding (e.g. those formed with prefixes like mis- or err-).

    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
     культуры
    -1.12
    喜欢
    -1.05
    eleven
    -1.04
     remor
    -1.02
    Institute
    -1.01
     Pedestrian
    -1.00
    콘
    -1.00
     bekannt
    -1.00
    voor
    -0.98
    確率
    -0.98
    POSITIVE LOGITS
     because
    1.60
     it
    1.41
     what
    1.37
     or
    1.20
     misunder
    1.19
     incorrectly
    1.16
     incorrect
    1.16
     often
    1.15
     expectations
    1.13
     неправи
    1.13
    Activations Density 0.045%

    No Known Activations