© 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. 24-RES-MATRYOSHKA-DC
    7. 80
    Prev
    Next
    INDEX
    Explanations

    capitalized words and phrases that appear to be commands or strong assertions

    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.24.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.24.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
    plaatst
    -0.39
    houding
    -0.36
    .~(\
    -0.35
    Fordítás
    -0.35
     ejus
    -0.35
    råd
    -0.34
    verlust
    -0.33
    Câu
    -0.33
     wortel
    -0.33
    öff
    -0.33
    POSITIVE LOGITS
     FOR
    0.50
     THE
    0.48
     IN
    0.48
     TO
    0.47
     INTO
    0.46
     THAN
    0.46
     WITH
    0.46
     ON
    0.45
     FROM
    0.45
     OVER
    0.45
    Activations Density 11.582%

    No Known Activations