© 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. 10-GEMMASCOPE-RES-131K
    4. 3843
    Prev
    Next
    INDEX
    Explanations

    who marked, at the, we will

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

    The neuron fires most strongly on open-class “content” tokens—nouns, verbs, numbers, question marks, and other semantically important words—rather than on 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
     that
    -1.73
     such
    -1.52
     you
    -1.45
     and
    -1.45
     primarily
    -1.43
     mainly
    -1.35
     about
    -1.32
     by
    -1.32
     create
    -1.30
     its
    -1.29
    POSITIVE LOGITS
    𖣔
    1.53
     beden
    1.42
    cách
    1.41
    pergillus
    1.40
    hoka
    1.37
     accentuated
    1.37
    rocław
    1.36
    粜
    1.35
    ágenes
    1.34
    ೬
    1.34
    Activations Density 0.125%

    No Known Activations