© 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. 61096
    Prev
    Next
    INDEX
    Explanations

    dead + description

    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
    åĥļ
    -0.11
    åķ¬
    -0.10
    泡泡
    -0.10
    ãģ£ãģįãĤĬ
    -0.10
    theses
    -0.09
    anders
    -0.09
    åħħ足
    -0.09
    çľ¼çļ®
    -0.09
    odate
    -0.08
     Tacoma
    -0.08
    POSITIVE LOGITS
    ness
    0.13
    liest
    0.12
    vertise
    0.11
    lier
    0.10
    liness
    0.09
    locking
    0.09
    (dead
    0.09
    æĶ¿åºľéĥ¨éŨ
    0.09
    Dead
    0.09
     dead
    0.09
    Activations Density 0.009%

    No Known Activations