© 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. 18-RES-MATRYOSHKA-DC
    7. 1158
    Prev
    Next
    INDEX
    Explanations

    words related to improving the efficiency or components of systems

    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.18.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.18.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
     صوتيه
    -0.85
     الحره
    -0.77
     continúas
    -0.70
     مشين
    -0.67
     محفوظة
    -0.67
     unknownFields
    -0.65
    AndEndTag
    -0.65
    rån
    -0.63
    WireFormatLite
    -0.61
    Архівовано
    -0.61
    POSITIVE LOGITS
     considerably
    1.08
     substantially
    0.97
     tremendously
    0.95
     significantly
    0.92
     greatly
    0.90
     drastically
    0.86
     dramatically
    0.86
     enormously
    0.85
     erheblich
    0.85
    significantly
    0.82
    Activations Density 1.450%

    No Known Activations