© 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. 105170
    Prev
    Next
    INDEX
    Explanations

    would + verb

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

    This neuron responds to common single‐word verbs (e.g. “look,” “go,” “take,” “do,” “say,” “make,” “get,” “happen”).

    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
    垰
    -2.48
    u
    -2.45
    茈
    -2.44
    轳
    -2.36
    
    -2.30
     protesta
    -2.25
    
    -2.17
     parlé
    -2.09
    
    -2.09
     comanda
    -2.05
    POSITIVE LOGITS
     Would
    2.44
    If
    2.41
     Since
    2.02
     This
    1.92
     With
    1.83
    是非常
    1.80
     Maybe
    1.80
    :
    1.78
     Perhaps
    1.78
     Either
    1.76
    Activations Density 0.034%

    No Known Activations