© 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-2-2B
    5. 22-RES-MATRYOSHKA-DC
    7. 1533
    Prev
    Next
    INDEX
    Explanations

    words related to political corruption, specifically leaking private information

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.22.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    principalTable
    -0.71
     MonoBehaviour
    -0.64
    جمعیت
    -0.58
     Référence
    -0.56
    endphp
    -0.55
    GNA
    -0.53
     jsPsych
    -0.53
    fører
    -0.52
    ArgsConstructor
    -0.52
    Greeting
    -0.52
    POSITIVE LOGITS
     secret
    1.51
     secrets
    1.37
     confidential
    1.32
     secrecy
    1.30
    secret
    1.27
     Secret
    1.24
    Secret
    1.23
     Geheim
    1.19
     SECRET
    1.18
    secrets
    1.18
    Activations Density 2.158%

    No Known Activations