© 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. 10-GEMMASCOPE-RES-131K
    7. 8382
    Prev
    Next
    INDEX
    Explanations

    how things happen

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

    This neuron fires on the main topic–defining words in section or document titles/headings.

    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
     who
    -1.69
    私が
    -1.33
    ρίου
    -1.30
    إنه
    -1.26
     vostra
    -1.26
     orilla
    -1.23
    erializer
    -1.19
    あなたの
    -1.19
    因为它
    -1.18
    </th>
    -1.17
    POSITIVE LOGITS
     different
    1.94
     each
    1.65
     שונים
    1.61
     poszczegól
    1.52
    ícias
    1.52
     various
    1.52
     diferentes
    1.45
     differs
    1.43
     varies
    1.41
     distintos
    1.41
    Activations Density 0.196%

    No Known Activations