© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49738
    Prev
    Next
    INDEX
    Explanations

    the + official titles

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

    The neuron activates on numeric identifiers and citations (e.g. section numbers, years, regulatory and docket codes).

    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
    tront
    -0.91
    ње
    -0.91
    ὦ
    -0.88
    ícias
    -0.87
    cèse
    -0.87
     crochet
    -0.86
    wrapper
    -0.86
    oghi
    -0.85
     golfers
    -0.84
    尭
    -0.84
    POSITIVE LOGITS
     Director
    2.16
     Secretary
    1.87
     President
    1.80
     director
    1.77
     Chief
    1.69
    Director
    1.66
     president
    1.57
     chief
    1.55
     Minister
    1.52
     CEO
    1.52
    Activations Density 0.066%

    No Known Activations