© 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. 93383
    Prev
    Next
    INDEX
    Explanations

    in this light

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

    The neuron is primarily detecting occurrences of the word “light.”

    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
     it
    -1.42
     in
    -1.41
     their
    -1.34
     you
    -1.27
     has
    -1.22
     very
    -1.13
     mainly
    -1.08
     my
    -1.05
     how
    -1.05
     they
    -1.04
    POSITIVE LOGITS
    identified
    1.09
    umenten
    1.08
     consentimento
    1.08
     クーポン
    1.06
    iscus
    1.05
     intellig
    1.02
     foglie
    1.02
    ﺮ
    1.01
    lored
    0.99
    졍
    0.99
    Activations Density 0.000%

    No Known Activations