© 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. 53554
    Prev
    Next
    INDEX
    Explanations

    lower, followed by descriptive terms

    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
    éĻħ
    -0.10
    ummer
    -0.10
    ually
    -0.09
     оÑĤз
    -0.09
    ä¸ŃéĢĶ
    -0.09
     ÙĪغÙĬر
    -0.09
    oris
    -0.08
    /from
    -0.08
    趣
    -0.08
    æĸ°æĺ¥
    -0.08
    POSITIVE LOGITS
    most
    0.16
    -most
    0.11
    MOST
    0.10
    halb
    0.10
    -than
    0.09
    ivent
    0.09
    åĩłçĤ¹
    0.09
    edImage
    0.09
    -middle
    0.09
    è´±
    0.09
    Activations Density 0.043%

    No Known Activations