© 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. 15624
    Prev
    Next
    INDEX
    Explanations

    UNITED STATES of America

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

    The neuron is keyed to fully uppercase words or acronyms (e.g. legal or technical headings and RUN labels).

    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
     trebui
    -1.15
     kullanılır
    -1.14
     szybko
    -1.12
    去找
    -1.09
     Empieza
    -1.09
    cestor
    -1.08
     ligera
    -1.08
     forestal
    -1.05
     veiligheid
    -1.05
     mientras
    -1.04
    POSITIVE LOGITS
     in
    1.12
     all
    1.12
     and
    1.03
    Lastly
    1.02
     only
    1.00
     ALL
    0.99
    endfor
    0.97
     Lastly
    0.96
    </em>
    0.95
     through
    0.94
    Activations Density 0.007%

    No Known Activations