© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 34851
    Prev
    Next
    INDEX
    Explanations

    instances and variations of the word "sneak" and its related forms, indicating stealthy behavior

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    sneak into, through, off

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WindowConstants
    -0.41
     what
    -0.40
     ור
    -0.40
     Issue
    -0.38
    ynb
    -0.38
     issues
    -0.37
    upaten
    -0.36
     onResponse
    -0.35
    ActionPerformed
    -0.35
     Issues
    -0.35
    POSITIVE LOGITS
     sneak
    1.12
    sneak
    1.09
     sneaking
    1.09
    Sneak
    1.07
     Sneak
    1.02
     sneaky
    0.90
     peek
    0.85
    Sneaky
    0.81
     stealth
    0.79
     crept
    0.75
    Activations Density 0.009%

    No Known Activations