© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10468
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts

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

    words or phrases related to specialized domains, systems, or technical features (such as royal estates, software frameworks, research programs, cosmetic formulas, project deadlines, technical directories, and competitive contexts).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron selectively fires on multi-syllable, content-bearing nouns (especially technical or formal terms).

    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
     of
    -2.00
     because
    -1.96
     like
    -1.69
     for
    -1.59
     you
    -1.58
     on
    -1.52
     inapropiados
    -1.52
     in
    -1.51
     who
    -1.49
     your
    -1.46
    POSITIVE LOGITS
     endeavours
    1.30
    孩子们
    1.27
     каждый
    1.23
     登录
    1.21
     endeavors
    1.21
    Ké
    1.18
     contin
    1.16
    ኋ
    1.14
     совершенно
    1.13
     condam
    1.12
    Activations Density 0.224%

    No Known Activations