© 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. 10-GEMMASCOPE-RES-131K
    7. 48089
    Prev
    Next
    INDEX
    Explanations

    __instrumentor_map__

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

    The neuron detects tokens made up of double‐underscore identifiers (i.e. names beginning with “__”).

    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
    ꒥
    -2.83
     misschien
    -2.27
    ientos
    -2.19
    ஂ
    -2.19
    硯
    -2.17
     Моск
    -2.14
    Cuándo
    -2.13
     років
    -2.09
    瑨
    -2.09
     Цер
    -2.06
    POSITIVE LOGITS
     was
    3.16
    2
    2.41
    }$
    2.38
    5
    2.36
    3
    2.36
    在他
    2.34
    寕
    2.30
    1
    2.28
     viendra
    2.27
    9
    2.23
    Activations Density 0.006%

    No Known Activations