© 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. 70920
    Prev
    Next
    INDEX
    Explanations

    intensifiers and strong adjectives

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

    The neuron detects highly emphatic evaluative words (strong adverbs or adjectives) used to intensify statements.

    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
     唯美
    -1.84
    ⽌
    -1.72
    foundland
    -1.71
    酲
    -1.70
    栒
    -1.66
    掟
    -1.63
     এবং
    -1.63
    lizes
    -1.62
    獃
    -1.62
    視野
    -1.61
    POSITIVE LOGITS
    ことも
    1.75
    Both
    1.68
    When
    1.63
    During
    1.51
     urging
    1.49
    為に
    1.48
    Although
    1.47
     müh
    1.47
    One
    1.46
    Within
    1.45
    Activations Density 0.060%

    No Known Activations