© 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. 35929
    Prev
    Next
    INDEX
    Explanations

    modifiers of actions or states

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

    The neuron detects adverbs—especially ones ending in “-ly”—that modify actions or states.

    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
     that
    -1.98
    などが
    -1.46
     these
    -1.42
     such
    -1.41
     about
    -1.38
     could
    -1.34
    '].
    -1.33
     might
    -1.30
    などは
    -1.28
     how
    -1.23
    POSITIVE LOGITS
     and
    1.68
     which
    1.34
     ürünleri
    1.31
    ѕ
    1.30
    のですか
    1.26
    ことができ
    1.26
    wendige
    1.26
    nemia
    1.24
    感を
    1.22
    venant
    1.21
    Activations Density 0.039%

    No Known Activations