© 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. 129701
    Prev
    Next
    INDEX
    Explanations

    The neuron detects strong action‐oriented verbs that describe functional or relational operations (e.g. directs, couples, generates, called out).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    actions and functions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     it
    -1.13
    endom
    -1.05
     становятся
    -0.99
     इसे
    -0.99
    ='';
    -0.93
    Several
    -0.93
    Your
    -0.93
     erneut
    -0.92
    Once
    -0.91
    这也是
    -0.91
    POSITIVE LOGITS
    esseurs
    1.16
     membantu
    1.15
     pomaga
    1.03
    primarily
    1.02
     klanten
    1.01
    看着
    1.00
     diğer
    0.98
    😒
    0.97
    strator
    0.96
     épaules
    0.96
    Activations Density 0.146%

    No Known Activations