© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127966
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on the stems of negative contractions (e.g. “don,” “didn,” “wasn,” etc.), i.e. occurrences of “…n’t.”

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

    negative contractions

    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
     jajaja
    -1.26
     länder
    -1.21
     oraz
    -1.21
    抡
    -1.16
     имму
    -1.13
     barcel
    -1.12
     departament
    -1.10
     each
    -1.09
     (
    -1.09
    fören
    -1.09
    POSITIVE LOGITS
     wasn
    1.46
     didn
    1.35
     isn
    1.31
    を行いました
    1.30
    を行います
    1.29
     rechazar
    1.27
    Conclusie
    1.24
    本日の
    1.20
     églises
    1.19
     doesn
    1.18
    Activations Density 0.003%

    No Known Activations