© 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. 22986
    Prev
    Next
    INDEX
    Explanations

    IN THE INTEREST OF

    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
    éĢĢ
    -0.10
    çĻ»
    -0.09
    çĺª
    -0.09
    æIJ½
    -0.09
    lime
    -0.08
    ä¾ĽåºĶéĵ¾
    -0.08
    rug
    -0.08
     гÑĢÑĥппÑĭ
    -0.08
    æķĿ
    -0.08
     trÃł
    -0.08
    POSITIVE LOGITS
    çĶ·æĸ¹
    0.12
    女æĸ¹
    0.12
     Couple
    0.10
    追赶
    0.09
    /forum
    0.09
    éĤķ
    0.09
    æĬļåħ»
    0.09
    ä¸Ģæĸ¹
    0.09
    æī¶
    0.09
     remar
    0.09
    Activations Density 0.090%

    No Known Activations