© 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. 92954
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by common words

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

    The neuron strongly activates on long, multi-syllable (often Latin-derived) words—e.g. technical or academic adjectives and participles.

    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
     gustado
    -1.53
     calendriers
    -1.52
    Langkah
    -1.42
     enlever
    -1.41
     apporter
    -1.33
    Karakteristik
    -1.30
     rembourse
    -1.30
    Berat
    -1.28
    Berikut
    -1.27
    Satu
    -1.27
    POSITIVE LOGITS
     of
    2.30
     from
    2.00
     for
    1.82
     on
    1.82
     to
    1.73
     that
    1.66
     in
    1.53
     between
    1.52
     about
    1.46
     at
    1.45
    Activations Density 0.051%

    No Known Activations