© 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. 2824
    Prev
    Next
    INDEX
    Explanations

    words and phrases typical of patent applications, especially those related to drilling and soil treatment.

    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
     proactively
    -0.87
     infrastruktur
    -0.78
     signifikan
    -0.78
     transitioning
    -0.75
     customizable
    -0.75
     setidaknya
    -0.74
     entornos
    -0.73
     onboarding
    -0.71
     seamlessly
    -0.70
     prioritized
    -0.70
    POSITIVE LOGITS
    faßt
    0.83
     Надо
    0.65
    Надо
    0.64
     convenable
    0.61
     mengadakan
    0.60
     daß
    0.60
     muß
    0.59
     läßt
    0.58
     Schluß
    0.58
     собою
    0.58
    Activations Density 1.306%

    No Known Activations