© 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. 127825
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on the word “duplicate” (in any casing), i.e. it detects occurrences of “duplicate.”

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    duplicate of or duplicate reference

    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
    Negative Logits
    ↵↵
    -2.53
     i
    -2.13
     `
    -2.11
    〬
    -2.08
     '
    -2.05
     profond
    -2.03
    \
    -2.02
    B
    -2.00
    みたら
    -1.96
     nurture
    -1.93
    POSITIVE LOGITS
    CurtirCurtir
    2.39
     camiset
    2.36
    以上です
    2.33
    ”،
    2.30
    adeira
    2.30
    蟶
    2.23
    衎
    2.23
    inerja
    2.22
    敺
    2.22
     parachoque
    2.19
    Activations Density 0.019%

    No Known Activations