© 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. 57998
    Prev
    Next
    INDEX
    Explanations

    but also

    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
    ãģ«ãĤĤ
    -0.10
    ä¸ĢåĽ¢
    -0.10
    ä¹ĭç¾İ
    -0.10
    ãģ¡ãĤĩãģĨãģ©
    -0.10
    ä¹ŁæŃ£æĺ¯
    -0.09
    cre
    -0.09
    mates
    -0.09
    ç¾İçĻ½
    -0.09
    åĢĴæĺ¯
    -0.09
     ÑĤакже
    -0.09
    POSITIVE LOGITS
    atten
    0.10
     ë¶Ī구íķĺê³ł
    0.10
    umber
    0.09
    åĵĩ
    0.09
    /or
    0.09
    åħ¼
    0.09
    vang
    0.09
    è¶Ł
    0.08
    ROID
    0.08
    äºĴ
    0.08
    Activations Density 0.020%

    No Known Activations