© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 38893
    Prev
    Next
    INDEX
    Explanations

    still state of being

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

    This neuron responds to verbs (especially simple past and third-person singular present forms).

    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
     about
    -1.53
     evangé
    -1.48
     ใหม่
    -1.41
    昻
    -1.41
    鮃
    -1.36
    .}(
    -1.34
     込
    -1.31
     pastillas
    -1.30
    ânica
    -1.30
     abbildung
    -1.29
    POSITIVE LOGITS
    仍
    1.41
     fortfarande
    1.30
     STILL
    1.28
     not
    1.27
     no
    1.25
     are
    1.23
     jovem
    1.19
     fortsatt
    1.19
    currently
    1.19
     still
    1.19
    Activations Density 0.052%

    No Known Activations