© 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. 106443
    Prev
    Next
    INDEX
    Explanations

    biz, baz, büz variations

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

    The neuron fires on any token containing the letter‐pair “bz” (e.g. in “bzip2,” “libbz2,” “Bazaar,” “Bizet,” etc.).

    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
    edrez
    -0.73
     nothing
    -0.72
     sophisticated
    -0.71
     something
    -0.70
     torino
    -0.68
     vnnd
    -0.68
     vnode
    -0.66
     pureza
    -0.65
     mengg
    -0.65
     Salinity
    -0.65
    POSITIVE LOGITS
    WTF
    0.77
     ovaries
    0.76
    autonomie
    0.75
    וך
    0.73
     USP
    0.73
    corner
    0.73
    恋
    0.71
     mitigating
    0.69
    Banyak
    0.69
    ís
    0.68
    Activations Density 0.021%

    No Known Activations