© 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. 68709
    Prev
    Next
    INDEX
    Explanations

    components and systems

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

    The neuron flags high-importance, domain-specific nouns (e.g. “company,” “shareholders,” “data,” “computer,” “environment”) that serve as topic keywords in technical or financial text.

    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
     and
    -1.59
     this
    -1.40
     then
    -1.24
     again
    -1.16
    ことも
    -1.15
    ところは
    -1.14
     dabar
    -1.05
     these
    -1.01
     asemenea
    -1.00
    際には
    -0.98
    POSITIVE LOGITS
    的一种
    1.70
    WillAppear
    1.37
     or
    1.27
    usually
    1.27
     aikana
    1.26
     is
    1.26
    段时间
    1.24
     who
    1.21
    において
    1.16
    欧式
    1.14
    Activations Density 0.140%

    No Known Activations