© 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. 110908
    Prev
    Next
    INDEX
    Explanations

    of followed by nouns

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

    The neuron responds strongly to process-oriented nominalizations (verbs turned into nouns, including gerunds)—especially in “X of Y” noun phrases.

    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
     baden
    -1.39
     frauen
    -1.36
    *
    -1.36
     wanda
    -1.36
     uruguay
    -1.32
     kien
    -1.28
     schli
    -1.27
     holz
    -1.27
     schal
    -1.27
     tisch
    -1.26
    POSITIVE LOGITS
     of
    2.36
     on
    1.59
     at
    1.55
     that
    1.55
     también
    1.42
     Dafür
    1.30
    更大的
    1.27
     historiador
    1.26
    少なくとも
    1.25
     bestehenden
    1.23
    Activations Density 0.224%

    No Known Activations