© 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. 38122
    Prev
    Next
    INDEX
    Explanations

    gentle actions following a preceding element

    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_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
    izas
    -0.81
    lüssel
    -0.79
    Refreshing
    -0.79
    thrown
    -0.78
     WALK
    -0.77
     démarches
    -0.77
    owała
    -0.76
    蜒
    -0.76
    してもら
    -0.74
    ctives
    -0.73
    POSITIVE LOGITS
     gently
    1.44
     kissing
    1.28
     kissed
    1.18
     kisses
    1.15
     pulling
    1.12
     pull
    1.07
     kiss
    1.07
     gentle
    1.05
     pulled
    1.02
     moved
    0.98
    Activations Density 0.005%

    No Known Activations