© 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. 64288
    Prev
    Next
    INDEX
    Explanations

    return item, provides access, view file, change value, select multiple

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

    The neuron detects Chinese‐language tokens (i.e. sequences of Chinese characters).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    <bos>
    -1.77
     to
    -1.74
    因为它
    -1.70
    us
    -1.57
    man
    -1.52
    os
    -1.51
    ist
    -1.46
    缇
    -1.37
    le
    -1.36
     is
    -1.31
    POSITIVE LOGITS
     banheiro
    2.02
     espet
    1.87
     our
    1.84
     salão
    1.71
     kokos
    1.71
     such
    1.69
     avião
    1.67
     the
    1.64
     farbe
    1.62
     cartaz
    1.61
    Activations Density 0.007%

    No Known Activations