© 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. 36625
    Prev
    Next
    INDEX
    Explanations

    hyper, Born, gluta, chlor, Bob, Th, Mar

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

    The neuron responds to subword pieces at the beginnings of longer, content-rich terms—especially technical or proper names—i.e. it flags the initial fragments (prefixes) of multi-syllable 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_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
     is
    -1.73
     in
    -1.56
     de
    -1.32
     to
    -1.32
     has
    -1.30
     
    -1.28
     B
    -1.27
     Y
    -1.25
     N
    -1.24
     are
    -1.22
    POSITIVE LOGITS
    }$.
    1.99
    }}$.
    1.73
    ។
    1.66
    ).
    1.61
    خخ
    1.55
     suav
    1.52
    ++.
    1.49
    ”.
    1.48
    kissen
    1.47
     ヌード
    1.45
    Activations Density 0.106%

    No Known Activations