© 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. 49382
    Prev
    Next
    INDEX
    Explanations

    permanently or adopted

    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
    ield
    -0.09
    åĪļ
    -0.08
     renewal
    -0.08
    ATED
    -0.08
     lắm
    -0.08
     decay
    -0.08
    oblin
    -0.08
    scrollTop
    -0.08
    viewer
    -0.08
     Renew
    -0.07
    POSITIVE LOGITS
    æŃ£å¼ı
    0.22
    æ°¸ä¹ħ
    0.19
     permanent
    0.18
     permanently
    0.16
    Permanent
    0.15
    permanent
    0.15
     perman
    0.14
    æŃ£å¼ıåIJ¯åĬ¨
    0.14
    æĹ¥æŃ£å¼ı
    0.14
     Permanent
    0.13
    Activations Density 0.370%

    No Known Activations