© 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. 70449
    Prev
    Next
    INDEX
    Explanations

    email subject lines

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

    The neuron responds to underscore characters—i.e. it fires on the “_” token (as used in identifiers or variable names).

    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
     to
    -1.57
    と聞
    -1.46
     enabling
    -1.42
     such
    -1.39
     "
    -1.36
     нього
    -1.36
    コミュ
    -1.35
     using
    -1.33
     また
    -1.33
     According
    -1.29
    POSITIVE LOGITS
    Ekster
    1.46
     rask
    1.45
    萏
    1.41
     Giappone
    1.40
     bruin
    1.38
    縒
    1.38
     '';
    1.33
    Réponse
    1.31
     avena
    1.31
    лению
    1.30
    Activations Density 0.007%

    No Known Activations