© 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. 67530
    Prev
    Next
    INDEX
    Explanations

    modifier noun compounds

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

    The neuron activates on descriptive words—mostly adjectives and adverbs—that signal magnitude, degree, or intensity (e.g. rapid, increased, high, dramatic, efficient).

    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
     mystère
    -1.17
     rhestr
    -1.09
    бол
    -1.09
    都已经
    -1.02
     territoire
    -1.01
    这类
    -0.99
     faſt
    -0.98
     podrían
    -0.98
    𝐙
    -0.98
    ᠂
    -0.98
    POSITIVE LOGITS
     a
    1.23
     our
    1.11
     will
    1.11
     approach
    1.08
     change
    1.07
     reducing
    1.06
     first
    1.05
    ждый
    1.01
     an
    1.00
     .....
    0.99
    Activations Density 0.079%

    No Known Activations