© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8649
    Prev
    Next
    INDEX
    Explanations

    afterwards

    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
    itud
    -0.11
     Blackburn
    -0.10
    istic
    -0.10
    ä¸Ĭå²Ĺ
    -0.09
    ÑĤоÑĢ
    -0.09
    worthy
    -0.09
    ä¸ĢæŶ
    -0.09
    pri
    -0.08
    座
    -0.08
    æ»ļ
    -0.08
    POSITIVE LOGITS
    ness
    0.11
    alse
    0.10
    éĩįè¦ģåĽłç´ł
    0.10
    ull
    0.09
    åħ¹
    0.09
    èĮ
    0.08
     Unidos
    0.08
    TestClass
    0.08
    rote
    0.08
    alan
    0.08
    Activations Density 0.012%

    No Known Activations