© 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. 34-GEMMASCOPE-RES-131K
    7. 45249
    Prev
    Next
    INDEX
    Explanations

    through or via

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

    The neuron strongly activates on key domain-oriented content nouns—especially plural nouns—that serve as the central topics or objects in a passage.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     eenige
    -1.27
    ANTES
    -1.12
    tinyos
    -1.11
    (!__
    -1.11
    бах
    -1.05
     jezt
    -1.03
     entidade
    -1.02
    苳
    -1.01
     géographie
    -1.00
     ぽ
    -1.00
    POSITIVE LOGITS
     through
    10.06
     через
    6.78
     via
    6.69
     melalui
    6.56
     através
    6.31
    through
    6.19
    Through
    6.19
     attraverso
    5.81
     Through
    5.72
    通过
    5.56
    Activations Density 1.183%

    No Known Activations