© 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. 34-GEMMASCOPE-RES-131K
    7. 23479
    Prev
    Next
    INDEX
    Explanations

    states of being or ongoing actions

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

    The neuron strongly activates on conjugated forms of the verb “to be” (is, are, was, were, etc.)—i.e. copula/auxiliary “be” verbs.

    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
     specialization
    -1.00
    FromNib
    -0.99
     ".";
    -0.98
     nuestro
    -0.98
     nordrhein
    -0.97
     mą
    -0.97
    ⅏
    -0.96
     испы
    -0.93
    häuse
    -0.93
    ographical
    -0.90
    POSITIVE LOGITS
     awaiting
    1.84
     waiting
    1.83
     planning
    1.77
     currently
    1.76
     working
    1.56
     considering
    1.55
     looking
    1.52
     expected
    1.51
     confident
    1.51
     hoping
    1.48
    Activations Density 0.055%

    No Known Activations