© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129226
    Prev
    Next
    INDEX
    Explanations

    dominate headlines, office, skies, air

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

    The neuron detects nouns naming arenas or “platforms” of prominence or dominance (e.g. headlines, airwaves, box office, skies).

    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
    lossenen
    -1.48
     personer
    -1.37
    rscheinlichkeit
    -1.34
     about
    -1.34
     osoba
    -1.29
    exels
    -1.26
    josé
    -1.24
    ními
    -1.23
    EXAMPLES
    -1.23
    roffene
    -1.23
    POSITIVE LOGITS
     from
    1.74
    of
    1.28
     with
    1.22
     از
    1.20
     depuis
    1.16
    แห่ง
    1.13
     desde
    1.09
    米国
    1.07
     große
    1.07
     J
    1.06
    Activations Density 0.084%

    No Known Activations