© 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. 22107
    Prev
    Next
    INDEX
    Explanations

    blender

    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
    eline
    -0.09
    _attention
    -0.08
    lijke
    -0.08
    Attention
    -0.08
     NotImplemented
    -0.08
    åıĺçİ°
    -0.08
    éĻIJå®ļ
    -0.08
    raction
    -0.08
    lijk
    -0.08
    าว
    -0.08
    POSITIVE LOGITS
    èľī
    0.10
    мед
    0.09
    ä¸Ģéĥ¨
    0.09
    é³IJ
    0.09
    iator
    0.09
    æѤæ¡Ī
    0.09
     pct
    0.09
    é£Ļ
    0.08
    æĤ¬å´ĸ
    0.08
    hare
    0.08
    Activations Density 0.009%

    No Known Activations