© 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. 69214
    Prev
    Next
    INDEX
    Explanations

    commitment to you

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

    The neuron strongly responds to very common function words (stopwords) like “we,” “you,” “and,” “to,” “the,” commas and periods—that is, common grammatical “glue” words.

    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
    θλη
    -1.03
     still
    -1.00
     先生
    -0.89
    如何在
    -0.89
    deu
    -0.88
     rağmen
    -0.88
    хівовано
    -0.87
    borderRadius
    -0.86
     Nom
    -0.84
    jelen
    -0.84
    POSITIVE LOGITS
     therefore
    1.68
     поэтому
    1.44
    そのため
    1.30
     accordingly
    1.18
     that
    1.18
     daarom
    1.12
    จึง
    1.09
    therefore
    1.08
     Aufführung
    1.05
     لذلك
    1.03
    Activations Density 0.060%

    No Known Activations