© 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. 10-GEMMASCOPE-RES-131K
    7. 75381
    Prev
    Next
    INDEX
    Explanations

    difficulty in defining or assessing

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

    The neuron detects meta‐linguistic action verbs used for evaluating, describing, defining, or comparing (e.g., compare, define, describe, review, answer).

    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
    ïc
    -1.27
    frequently
    -1.20
     EVEN
    -1.14
    >
    -1.13
    similar
    -1.10
    нимание
    -1.10
    $};
    -1.09
     často
    -1.09
     interessante
    -1.08
    TOO
    -1.08
    POSITIVE LOGITS
     then
    1.48
     it
    1.45
     because
    1.37
     now
    1.36
     different
    1.34
     from
    1.27
     on
    1.27
     since
    1.27
     people
    1.27
     about
    1.18
    Activations Density 0.096%

    No Known Activations