© 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. 21573
    Prev
    Next
    INDEX
    Explanations

    crazy, mad, mental states

    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
    æĽ´å¤ļä¿¡æģ¯
    -0.09
    æºĺ
    -0.09
    col
    -0.09
    èµ·æŃ¥
    -0.09
    iable
    -0.09
    åħ¥çĿ¡
    -0.08
    åĢı
    -0.08
    Copying
    -0.08
     slump
    -0.08
    isphere
    -0.08
    POSITIVE LOGITS
    为æѤ
    0.11
    ç»į
    0.10
    为客æĪ·
    0.10
    çķ´
    0.10
    羡
    0.10
    ç¹ģèį£
    0.09
    æľīæķĪæľŁ
    0.09
    ãģ¶ãĤĬ
    0.09
    éķ¶
    0.09
     оÑĤд
    0.08
    Activations Density 0.020%

    No Known Activations