© 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. 49834
    Prev
    Next
    INDEX
    Explanations

    heartbreaker, broken heart, heartfelt feelings, breathtaking scenery

    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
    erg
    -0.09
    çıį
    -0.09
    lotte
    -0.08
    iment
    -0.08
    мол
    -0.08
    mtree
    -0.08
    çĸĹ
    -0.08
    topic
    -0.08
    æĸĩä½ĵ
    -0.08
    欢快
    -0.08
    POSITIVE LOGITS
    èı½
    0.09
    omic
    0.09
    éĵĦ
    0.09
    ä¹ĭä½Ļ
    0.09
    ved
    0.08
    çĬ¯è§Ħ
    0.08
    ening
    0.08
     Dock
    0.08
    agements
    0.08
    quee
    0.08
    Activations Density 0.007%

    No Known Activations