© 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. 130466
    Prev
    Next
    INDEX
    Explanations

    going along or out

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

    The neuron responds to the particle words (like “along,” “out,” “coming”) that form part of phrasal verbs indicating movement or participation.

    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
     that
    -2.06
     ies
    -1.45
     Heav
    -1.40
    
    -1.38
     indiquer
    -1.38
    鸸
    -1.34
     Sociale
    -1.33
     zijne
    -1.33
    
    -1.31
    
    -1.30
    POSITIVE LOGITS
     are
    1.73
     see
    1.58
     to
    1.53
     for
    1.52
     '
    1.48
    )
    1.46
     of
    1.45
     one
    1.42
     as
    1.39
     can
    1.38
    Activations Density 0.014%

    No Known Activations