© 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. 22-GEMMASCOPE-RES-131K
    7. 10897
    Prev
    Next
    INDEX
    Explanations

    advancing/improving/changing states

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

    The neuron fires on high‐frequency present‐tense auxiliary or linking verbs (especially “are,” plus related forms like “being,” “have,” “can,” etc.) used in generic statements.

    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
     when
    -1.66
     our
    -1.40
     what
    -1.38
     people
    -1.38
     before
    -1.36
     one
    -1.33
     as
    -1.33
     which
    -1.28
     after
    -1.28
     if
    -1.27
    POSITIVE LOGITS
    
    1.37
    还能
    1.27
    basically
    1.25
     mejorando
    1.22
     aumentando
    1.22
    generally
    1.22
     themselves
    1.21
    began
    1.20
     cambiando
    1.19
     colocando
    1.19
    Activations Density 0.067%

    No Known Activations