© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 78642
    Prev
    Next
    INDEX
    Explanations

    qualities of concepts

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

    The neuron fires on descriptive adjectives—words that characterize qualities or states (e.g. “struggling,” “slippery,” “tremendous,” “predictable,” “complex”).

    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
    什么呢
    -1.41
     péro
    -1.29
     brillantes
    -1.27
    𓁹
    -1.26
     kolik
    -1.26
    
    -1.24
     neuve
    -1.23
     cabelo
    -1.23
    僰
    -1.23
     novin
    -1.21
    POSITIVE LOGITS
     and
    1.95
    ness
    1.73
     nature
    1.67
     aspects
    1.66
     but
    1.53
     aspect
    1.36
     increase
    1.34
     των
    1.34
     of
    1.30
     amount
    1.28
    Activations Density 0.264%

    No Known Activations