© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 40136
    Prev
    Next
    INDEX
    Explanations

    evaluative adjectives preceding nouns

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

    The neuron fires on strong positive evaluative adjectives and intensifiers that highlight or praise qualities (e.g. “bad-ass,” “elegant,” “great,” “rich,” “exciting,” etc.).

    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
    ensitive
    -1.41
     Municipalidad
    -1.40
    に関しては
    -1.38
     Otras
    -1.35
    などを
    -1.34
     بالنسبة
    -1.32
     Cómo
    -1.32
     Valores
    -1.29
     Imágenes
    -1.28
     oth
    -1.26
    POSITIVE LOGITS
     and
    1.41
     what
    1.40
     this
    1.38
     their
    1.34
     our
    1.34
     both
    1.29
     having
    1.28
     creating
    1.28
     there
    1.27
     which
    1.24
    Activations Density 0.117%

    No Known Activations