© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 30-GEMMASCOPE-RES-16K
    6. 15031
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to legal decisions and agreements

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

    Decision to not do something

    np_acts-logits-general · gemini-2.0-flash

    refusal or inaction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uxxxx
    -0.37
     revi
    -0.36
    isSuccess
    -0.34
    シス
    -0.31
    あく
    -0.31
     rega
    -0.31
    slidesToShow
    -0.31
     Fati
    -0.31
     reorgan
    -0.31
     BoxDecoration
    -0.30
    POSITIVE LOGITS
     refusal
    1.28
     عدم
    1.26
     refusing
    1.21
     abstain
    1.21
     omission
    1.20
     absence
    1.11
     refused
    1.08
     absten
    1.07
     отсутствие
    1.07
     refus
    1.05
    Activations Density 1.733%

    No Known Activations