© 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. 32554
    Prev
    Next
    INDEX
    Explanations

    preference

    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
    æľįåĬ¡è´¨éĩı
    -0.12
    icker
    -0.09
    ifying
    -0.09
    د
    -0.09
    zza
    -0.09
    æľįåĬ¡æ°´å¹³
    -0.09
    el
    -0.09
    gest
    -0.09
    jes
    -0.08
    æľįåĬ¡èĥ½åĬĽ
    -0.08
    POSITIVE LOGITS
    ably
    0.18
    æĹ¶æľº
    0.11
    /op
    0.09
    ë§ģ
    0.09
    å®ĵ
    0.09
    ableObject
    0.09
    äºijéĽĨ
    0.09
    few
    0.09
    Desk
    0.08
    ç¾¤å²Ľ
    0.08
    Activations Density 0.014%

    No Known Activations