© 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. 54991
    Prev
    Next
    INDEX
    Explanations

    door or window handles

    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
    æ¸ħæĸ°
    -0.09
    æ¾Ħæ¸ħ
    -0.09
    æī©
    -0.09
    ood
    -0.09
    çĸ£
    -0.08
    _TC
    -0.08
    å±ģ
    -0.08
    å¹Ħ
    -0.08
     Lay
    -0.08
    æĴŃ
    -0.08
    POSITIVE LOGITS
    ajar
    0.11
    öff
    0.10
    äºĶéĩij
    0.10
    å³Ļ
    0.10
     slammed
    0.10
    èģĶ
    0.10
    .getRaw
    0.09
    .getWorld
    0.09
    åĿ¯
    0.09
     Unlock
    0.09
    Activations Density 0.175%

    No Known Activations