© 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. 22-GEMMASCOPE-RES-131K
    7. 67045
    Prev
    Next
    INDEX
    Explanations

    descriptive modifier then noun

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

    The neuron detects descriptive qualifier words—especially informative adjectives or adjective‐like modifiers—used to characterize or qualify nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     และ
    -2.27
     quintessential
    -2.16
    蟥
    -2.06
    
    -2.05
     trois
    -2.05
     momentous
    -2.05
     acclaimed
    -2.03
    ſſen
    -2.00
    
    -1.99
     startet
    -1.98
    POSITIVE LOGITS
     what
    2.22
    There
    2.13
     During
    2.11
     undeniably
    2.09
     or
    2.09
    Even
    2.06
     facilitating
    2.03
     horrific
    2.03
    With
    1.99
     supremely
    1.99
    Activations Density 0.236%

    No Known Activations