© 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. 10-GEMMASCOPE-RES-131K
    4. 105030
    Prev
    Next
    INDEX
    Explanations

    explaining why or the issue

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

    The neuron fires on language that flags something as important or significant (e.g. “very important,” “raises a point,” “has become increasingly important”).

    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
     this
    -1.79
     Wartawan
    -1.43
     Arxiv
    -1.38
    wwww
    -1.37
    wwwww
    -1.36
     these
    -1.35
     XXXI
    -1.35
     ©️
    -1.33
     XXIX
    -1.32
     XXXII
    -1.32
    POSITIVE LOGITS
     -
    1.52
     why
    1.35
    以后
    1.30
     *
    1.18
     :
    1.16
     ;
    1.16
     показать
    1.15
    Gesellschaft
    1.15
    全世界
    1.15
    ATING
    1.15
    Activations Density 0.059%

    No Known Activations