© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 124618
    Prev
    Next
    INDEX
    Explanations

    interplay between

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

    The neuron responds to words that signal an interaction or evocation of ideas (for example, verbs that “bring something up” and nouns denoting relationships between things).

    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
     leaſt
    -2.58
     we
    -2.41
    you
    -2.19
     ")");
    -2.13
    ۔
    -2.08
    ):
    -2.06
    ='';
    -2.06
     bisschen
    -1.99
    man
    -1.98
    謦
    -1.98
    POSITIVE LOGITS
    
    2.38
     because
    2.16
    Once
    2.06
    寕
    2.06
     There
    2.02
    Furthermore
    1.94
     voegen
    1.94
    𐄙
    1.92
    Also
    1.91
    vsk
    1.89
    Activations Density 0.003%

    No Known Activations