© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 53123
    Prev
    Next
    INDEX
    Explanations

    exactly what

    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
    æĺ¯ä½ł
    -0.09
    ç»·
    -0.09
    è°ģçŁ¥éģĵ
    -0.09
    omics
    -0.09
    weed
    -0.09
    èĥ½æĬĬ
    -0.09
    æģĥ
    -0.08
    caret
    -0.08
    ìłľ
    -0.08
    æĢ»èµĦ产
    -0.08
    POSITIVE LOGITS
    ocker
    0.09
    å±ĢéĿ¢
    0.09
    æijĨåľ¨
    0.08
    _ORIENTATION
    0.08
    .decorators
    0.08
    orts
    0.08
    äºĴåĬ©
    0.08
    åĮĸ
    0.08
     Eden
    0.08
    osten
    0.08
    Activations Density 0.020%

    No Known Activations