© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10424
    Prev
    Next
    INDEX
    Explanations

    caught up in

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

    The neuron responds strongly to the phrase “caught up,” i.e. the phrasal verb “caught up (in…).”

    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
    '
    -2.33
    Its
    -1.91
     when
    -1.90
     is
    -1.82
    It
    -1.77
    )
    -1.68
    when
    -1.64
     quando
    -1.63
    This
    -1.63
     from
    -1.63
    POSITIVE LOGITS
    閎
    1.83
    Considerations
    1.79
    Métodos
    1.74
    ⑊
    1.73
    妧
    1.70
    нология
    1.66
    
    1.66
    inerary
    1.65
     about
    1.64
    íticos
    1.64
    Activations Density 0.012%

    No Known Activations