© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28070
    Prev
    Next
    INDEX
    Explanations

    symbols, special characters, etc

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

    The neuron fires strongly on single non-alphanumeric symbols (punctuation or math symbols) and accented/extended characters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    になりたい
    -1.38
    bruno
    -1.38
    の方も
    -1.35
    にします
    -1.34
    にしました
    -1.32
     needed
    -1.30
    為に
    -1.28
    の方法
    -1.26
    づけ
    -1.24
    の方に
    -1.22
    POSITIVE LOGITS
    <bos>
    1.67
    などの
    1.55
    外部連結
    1.37
    そういった
    1.31
     quartiere
    1.27
     etc
    1.23
    edoria
    1.23
    excluding
    1.21
     canale
    1.20
    曲は
    1.20
    Activations Density 0.009%

    No Known Activations