© 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. 54043
    Prev
    Next
    INDEX
    Explanations

    classic book/movie/track

    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
    èµ·çłģ
    -0.09
    éĤ£ç§į
    -0.09
    éºĿ
    -0.09
     procur
    -0.08
    éĤ£ä½į
    -0.08
    ogn
    -0.08
    主人
    -0.08
    èĨ
    -0.08
    å§ĵ
    -0.08
    uttle
    -0.08
    POSITIVE LOGITS
    ä¼ļè®©ä½ł
    0.11
    åĽłæŀľ
    0.09
     акÑĤÑĥ
    0.09
    两年åīį
    0.09
    licable
    0.09
    redo
    0.09
    æݨåĩºäºĨ
    0.09
    æ§Ľ
    0.09
    egl
    0.08
    DidLoad
    0.08
    Activations Density 0.075%

    No Known Activations