© 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. 130935
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on the noun “reversal” (i.e. words naming the undoing of a process).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    reversal of

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     涂
    -0.97
    ૮
    -0.95
    downloader
    -0.95
     preventivo
    -0.94
    standigheden
    -0.92
    jupe
    -0.92
    urnya
    -0.91
     Untersuch
    -0.91
    ತ
    -0.91
     آهنگ
    -0.88
    POSITIVE LOGITS
     precedenti
    1.09
    menistan
    1.02
    volve
    0.98
    .*")]
    0.98
    ube
    0.97
    UPAC
    0.93
     détruire
    0.93
    ifrance
    0.93
    小声
    0.93
     Архивировано
    0.92
    Activations Density 0.000%

    No Known Activations