© 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. 1479
    Prev
    Next
    INDEX
    Explanations

    words related to social and political issues, but not consistently.

    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
     have
    -0.77
    hésitez
    -0.71
     require
    -0.71
     devrez
    -0.71
     rely
    -0.69
     deserve
    -0.69
     remain
    -0.69
     serve
    -0.68
     depend
    -0.68
     symbolize
    -0.67
    POSITIVE LOGITS
    gets
    0.73
     buys
    0.70
     pulls
    0.69
     tries
    0.69
     picks
    0.69
     pours
    0.69
     learns
    0.69
     washes
    0.69
     returns
    0.68
     drives
    0.68
    Activations Density 1.543%

    No Known Activations