© 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. 34-GEMMASCOPE-RES-131K
    7. 107718
    Prev
    Next
    INDEX
    Explanations

    UV photo; all relevant; A few; Q:

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

    This neuron activates strongly on title‐and‐headline text, essentially picking out words that occur in news article headlines or section titles.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Å
    -1.05
     hâ
    -1.03
     engend
    -1.02
    户外
    -1.02
     medy
    -1.02
    Anybody
    -1.02
     Í
    -1.00
    somebody
    -0.98
     kred
    -0.98
     trak
    -0.98
    POSITIVE LOGITS
     including
    1.45
     only
    1.08
     there
    1.03
     includes
    1.00
     directly
    0.99
     notare
    0.99
     beneficial
    0.98
     included
    0.98
     important
    0.98
     everyone
    0.98
    Activations Density 0.004%

    No Known Activations