© 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. 22-GEMMASCOPE-RES-131K
    7. 129694
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on capitalized words or tokens—especially proper names and sentence-starts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Filipino currency and cultural references

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     katun
    -2.00
     susun
    -1.74
     iyon
    -1.59
     makita
    -1.55
     aktar
    -1.41
     sapat
    -1.38
     siyang
    -1.38
     loob
    -1.38
     akong
    -1.33
     hina
    -1.31
    POSITIVE LOGITS
     their
    1.29
    [
    1.22
     šal
    1.20
     its
    1.17
    ₱
    1.16
    Zeneca
    1.14
    رول
    1.13
     supposedly
    1.11
     dapat
    1.09
     but
    1.08
    Activations Density 0.302%

    No Known Activations