© 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. 10-GEMMASCOPE-RES-131K
    7. 9046
    Prev
    Next
    INDEX
    Explanations

    names and legal cases

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

    The neuron fires on numeric tokens—especially the numbers in legal citations (volumes, years, page numbers, 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_10/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
    </em>
    -1.82
    for
    -1.73
     -
    -1.66
    </strong>
    -1.64
    ]
    -1.59
    </i>
    -1.54
     after
    -1.52
     very
    -1.49
    </u>
    -1.48
     if
    -1.45
    POSITIVE LOGITS
    oprene
    1.75
    zione
    1.68
    梠
    1.58
    grazie
    1.52
     desarrollo
    1.51
    ড়
    1.50
     convivial
    1.48
     appeler
    1.47
    gráficas
    1.45
    
    1.44
    Activations Density 0.011%

    No Known Activations