© 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. 34-GEMMASCOPE-RES-131K
    4. 116779
    Prev
    Next
    INDEX
    Explanations

    lists joined by and

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

    The neuron flags content-bearing words—mostly nouns and adjectives naming topics, domains, or salient concepts—while ignoring grammatical “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_34/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
    ;
    -2.31
     because
    -1.80
     in
    -1.77
    :
    -1.74
     which
    -1.73
    ).
    -1.65
    .
    -1.59
     that
    -1.58
    ),
    -1.56
     whose
    -1.55
    POSITIVE LOGITS
     and
    2.52
     etc
    1.97
     samt
    1.40
     valamint
    1.39
     eenen
    1.38
    kapital
    1.38
    ictwa
    1.36
     sowie
    1.31
     eenige
    1.28
     kapak
    1.27
    Activations Density 0.201%

    No Known Activations