© 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. 34-GEMMASCOPE-RES-131K
    7. 118698
    Prev
    Next
    INDEX
    Explanations

    proprietary licenses, code, or dates

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

    The neuron fires strongly on tokens composed of digits (e.g. version numbers, years, or other numeric strings).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     participant
    -0.85
     Roosevelt
    -0.74
    participant
    -0.71
     module
    -0.69
     tool
    -0.69
    Morrison
    -0.69
     prepared
    -0.68
    oose
    -0.68
    pulos
    -0.68
    ecos
    -0.66
    POSITIVE LOGITS
     Siddi
    0.75
    MEK
    0.72
    eline
    0.67
     sauvegarde
    0.65
     russia
    0.64
    Feb
    0.63
    akterysty
    0.63
    Примечания
    0.63
    сков
    0.62
     RUSS
    0.62
    Activations Density 0.108%

    No Known Activations