© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 2637
    Prev
    Next
    INDEX
    Explanations

    onboard or aboard

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oki
    -0.09
    æİĴéĺŁ
    -0.09
    SPORT
    -0.09
    otte
    -0.09
    ÐŀÐł
    -0.09
    otti
    -0.09
    places
    -0.08
     ÑĤоÑĢгов
    -0.08
    ROY
    -0.08
    æģ«
    -0.08
    POSITIVE LOGITS
     Shan
    0.09
     dash
    0.09
    诣
    0.09
    çĥŃæIJľ
    0.08
    /off
    0.08
    馨
    0.08
     ru
    0.08
    åįī
    0.08
    _DEFIN
    0.08
    å¤ĩ
    0.08
    Activations Density 0.031%

    No Known Activations