© 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. 22-GEMMASCOPE-RES-131K
    7. 72028
    Prev
    Next
    INDEX
    Explanations

    pronouns followed by action verbs

    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
     we
    -1.65
    laufen
    -1.10
    beca
    -1.09
    大大
    -1.02
    borderFill
    -1.02
     早餐
    -1.01
    ggy
    -1.01
    ,//
    -0.98
     Væ
    -0.97
    кож
    -0.95
    POSITIVE LOGITS
     knew
    1.83
     could
    1.58
     know
    1.56
     still
    1.55
     would
    1.44
     hadn
    1.43
     certainly
    1.42
     wanted
    1.41
     understand
    1.41
     think
    1.39
    Activations Density 0.005%

    No Known Activations