© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 93478
    Prev
    Next
    INDEX
    Explanations

    PRWEB dateline identifiers

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

    The neuron detects uppercase press‐release or document boilerplate markers—things like PRWEB, SBWIRE, copyright notices, and similar all-caps metadata/separator tokens.

    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
     butterknife
    -0.87
    Tak
    -0.87
    ネマ
    -0.86
    慢
    -0.84
     väg
    -0.79
    ро
    -0.77
     männis
    -0.77
     Traditions
    -0.77
     dak
    -0.76
    ==(
    -0.76
    POSITIVE LOGITS
    irmed
    0.85
     late
    0.80
     behind
    0.79
     kiedyś
    0.78
     asked
    0.77
    Estudios
    0.77
     verfü
    0.76
    仕方ない
    0.75
    馬鹿
    0.75
     outside
    0.74
    Activations Density 0.007%

    No Known Activations