© 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. 111025
    Prev
    Next
    INDEX
    Explanations

    quantifiers and descriptors

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

    The neuron fires strongly on relatively rare, content‐bearing words (i.e. high–tf–idf keywords) rather than common function 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_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
     will
    -1.44
     opět
    -1.41
     Tujuan
    -1.39
    chte
    -1.34
     کمتر
    -1.32
     autre
    -1.31
     nawet
    -1.28
     حداقل
    -1.27
    handleSubmit
    -1.26
     Seperti
    -1.24
    POSITIVE LOGITS
     on
    1.54
     things
    1.41
     erwäh
    1.36
     the
    1.34
     a
    1.33
     anspruchs
    1.32
    iocene
    1.30
     wysokiej
    1.28
    lijden
    1.27
    celand
    1.25
    Activations Density 0.534%

    No Known Activations