© 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 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. 80071
    Prev
    Next
    INDEX
    Explanations

    references to leaks or leaking information

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

    leak or leakage of information

    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
     CommandType
    -0.54
     functools
    -0.52
     numberWith
    -0.50
    opsida
    -0.49
    ptonshire
    -0.47
    CommandType
    -0.47
    atoday
    -0.47
    RectangleBorder
    -0.45
     Bons
    -0.44
    msen
    -0.44
    POSITIVE LOGITS
     leak
    1.16
    Leak
    1.14
     Leak
    1.14
    leak
    1.13
     leaks
    1.03
     Leaks
    0.96
     Leakage
    0.94
     leakage
    0.93
     leaking
    0.91
     leaked
    0.88
    Activations Density 0.009%

    No Known Activations