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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33368
    Prev
    Next
    INDEX
    Explanations

    cities and places

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

    This neuron activates on all-uppercase letter sequences (e.g. acronyms or words written entirely in capital 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
     other
    -1.38
     after
    -1.24
     where
    -1.20
     on
    -1.13
     through
    -1.12
     around
    -1.11
     with
    -1.10
     which
    -1.09
     some
    -1.09
     at
    -1.09
    POSITIVE LOGITS
     hwnd
    1.04
     remarkably
    1.02
     strikingly
    0.99
     ترجمه
    0.98
     delightfully
    0.97
     DARK
    0.97
     WRIGHT
    0.96
    牙齿
    0.96
     sâ
    0.96
     decidedly
    0.96
    Activations Density 0.004%

    No Known Activations