© 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. 38784
    Prev
    Next
    INDEX
    Explanations

    powder and meal ingredients

    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
    Õ¢
    -0.10
    éĶ°
    -0.10
    æ²»çĹħ
    -0.10
    çĸ¾çĹħçļĦ
    -0.10
    æł¹æľ¬ä¸į
    -0.09
    ä¸įæĺİçĻ½
    -0.09
    çIJĨ工大åѦ
    -0.09
    à¹ģà¸ĭ
    -0.09
    allis
    -0.09
    åıįåĵį
    -0.08
    POSITIVE LOGITS
     ÑĦоÑĢ
    0.10
    éľŃ
    0.09
    rado
    0.08
    峡
    0.08
    æī§çĿĢ
    0.08
    åĽ½å®¶æłĩåĩĨ
    0.08
    第ä¸Ģ个
    0.08
    inch
    0.08
    ä¸Ŀä¸Ŀ
    0.08
    vice
    0.08
    Activations Density 0.008%

    No Known Activations