© 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. 4358
    Prev
    Next
    INDEX
    Explanations

    in its entirety

    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
    èĻĵ
    -0.10
    hatt
    -0.10
    _frontend
    -0.10
    æļij
    -0.09
     DISPATCH
    -0.09
    æĭĬ
    -0.09
    _EXTENSIONS
    -0.09
    åīĭ
    -0.09
    åľ¨å¤ĸéĿ¢
    -0.09
    roperties
    -0.09
    POSITIVE LOGITS
     entirety
    0.18
    å®Įæķ´çļĦ
    0.17
    å®Įæķ´
    0.17
     entire
    0.15
     whole
    0.15
     intact
    0.14
    whole
    0.14
    åħ¨
    0.13
    åħ¨æĸĩ
    0.13
     toÃłn
    0.13
    Activations Density 0.105%

    No Known Activations