© 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. 129399
    Prev
    Next
    INDEX
    Explanations

    foreign language words

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

    The main thing this neuron does is detect ALL-CAPS words or phrases used for shouting or emphasis.

    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
    5
    -2.02
    6
    -1.71
     Depending
    -1.69
    ].
    -1.66
    7
    -1.55
     your
    -1.54
    If
    -1.54
     with
    -1.54
    ][
    -1.54
     sonriente
    -1.53
    POSITIVE LOGITS
    踯
    1.80
    側は
    1.77
    犼
    1.70
    циях
    1.67
    lovakia
    1.66
     vijf
    1.62
    を活用
    1.61
     nunmehr
    1.59
     zwölf
    1.59
     vanguardia
    1.59
    Activations Density 0.059%

    No Known Activations