© 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. 14309
    Prev
    Next
    INDEX
    Explanations

    first part of names and compound words

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

    The neuron activates on capitalized tokens representing named entities (proper nouns), such as people’s names, organization names, or other proper nouns.

    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
     our
    -1.79
     ТАК
    -1.38
     recientemente
    -1.35
     recientes
    -1.29
     всички
    -1.28
     fuß
    -1.25
     unseren
    -1.23
     reír
    -1.23
     laiton
    -1.22
     permissible
    -1.21
    POSITIVE LOGITS
     like
    1.71
    ⚣
    1.61
    signable
    1.57
    yship
    1.52
    criminate
    1.48
     wasn
    1.47
    uminum
    1.45
     ſta
    1.44
    seteq
    1.43
    essment
    1.38
    Activations Density 0.006%

    No Known Activations