© 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. 23205
    Prev
    Next
    INDEX
    Explanations

    recipes

    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
    aker
    -0.09
    éķ°
    -0.09
    listed
    -0.09
    ArrayOf
    -0.09
    mass
    -0.08
    ogh
    -0.08
    ning
    -0.08
    aji
    -0.08
    tat
    -0.08
    ingly
    -0.08
    POSITIVE LOGITS
    大åħ¨
    0.10
    -holder
    0.10
    AndServe
    0.10
    çİĭåĽ½
    0.10
    çķ´
    0.09
    uled
    0.09
    æĿ¿æĿIJ
    0.09
    ä¹İ
    0.09
    åĬłçĽĸ
    0.09
    çŃīå½¢å¼ı
    0.09
    Activations Density 0.007%

    No Known Activations