© 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. 10-GEMMASCOPE-RES-131K
    4. 41923
    Prev
    Next
    INDEX
    Explanations

    seal, seals, sealed

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

    The main thing this neuron does is detect technical terms related to seals and sealing (e.g., “seal,” “sealing,” “sealed”) in mechanical or engineering contexts.

    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
    dáms
    -1.38
    廨
    -1.34
    籏
    -1.31
     umožňuje
    -1.30
    ઁ
    -1.30
     現貨
    -1.29
     nabízí
    -1.28
    注意事
    -1.27
     Hinblick
    -1.25
    炱
    -1.24
    POSITIVE LOGITS
     that
    1.42
     which
    1.33
    」
    1.27
    oak
    1.26
    4
    1.21
     experience
    1.19
     he
    1.18
    ="">
    1.16
    ”.
    1.12
     contact
    1.10
    Activations Density 0.059%

    No Known Activations