© 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. 50121
    Prev
    Next
    INDEX
    Explanations

    switch between modes

    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
    <dd
    -0.10
    çĸı
    -0.09
    åĻª
    -0.09
    .collect
    -0.08
    STALL
    -0.08
    SSIP
    -0.08
     Sims
    -0.08
    interactive
    -0.08
    collect
    -0.08
    äºĵ
    -0.08
    POSITIVE LOGITS
    åĪĩæį¢
    0.28
     switch
    0.19
     switching
    0.19
    switch
    0.17
    Switch
    0.17
     Switch
    0.16
    _switch
    0.16
    -switch
    0.15
    	switch
    0.15
     switches
    0.15
    Activations Density 0.384%

    No Known Activations