© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66805
    Prev
    Next
    INDEX
    Explanations

    Goshen, Mie, We, Q

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

    The neuron detects non‐ASCII Unicode symbols—especially typographic or accented characters (e.g. ©, ®, ™, °, †, ¶, accented letters).

    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
     it
    -1.41
     somber
    -1.23
    kyrie
    -1.22
     peixe
    -1.20
    ඵ
    -1.18
    ApJ
    -1.12
     dadas
    -1.11
     önemli
    -1.10
     ovos
    -1.09
    🎒
    -1.09
    POSITIVE LOGITS
    も
    1.24
    recently
    1.21
     Karena
    1.20
    その
    1.16
     drücken
    1.13
    still
    1.12
    更好的
    1.09
    following
    1.08
     несколько
    1.07
    与此同时
    1.07
    Activations Density 0.000%

    No Known Activations