© 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. 68708
    Prev
    Next
    INDEX
    Explanations

    criticism or banning exchanges

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

    The neuron activates on words written in title case—especially proper names and headline-style capitalized tokens.

    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
     will
    -1.50
     without
    -1.26
     at
    -1.14
    )
    -1.11
     as
    -1.11
    剧照
    -1.10
     other
    -1.09
     so
    -1.09
     As
    -1.09
    お試し
    -1.05
    POSITIVE LOGITS
     respetu
    1.38
    alucía
    1.30
    黉
    1.28
     reager
    1.27
     joyería
    1.22
    menü
    1.20
    indirect
    1.17
     verschillende
    1.16
    kompet
    1.16
    ---->
    1.13
    Activations Density 0.050%

    No Known Activations