© 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. 6328
    Prev
    Next
    INDEX
    Explanations

    wall or walls

    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
    çĥŁ
    -0.10
    éĿ¢çĽ®
    -0.10
    (DBG
    -0.09
    横
    -0.09
    elpers
    -0.09
    eração
    -0.09
    iban
    -0.08
    казаÑĤÑĮ
    -0.08
    illance
    -0.08
    ìķ½
    -0.08
    POSITIVE LOGITS
    pan
    0.10
    ä¸ĬçļĦ
    0.09
    dest
    0.09
    çļĦä¸įåIJĮ
    0.09
    ional
    0.09
    éļĻ
    0.08
     PAN
    0.08
     boÅŁ
    0.08
    ç¬ijè¯Ŀ
    0.08
     dest
    0.08
    Activations Density 0.040%

    No Known Activations