© 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. 27346
    Prev
    Next
    INDEX
    Explanations

    really want

    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
    å·²ä¹ħçļĦ
    -0.10
    åŁºæľ¬åİŁåĪĻ
    -0.09
    åĵĤ
    -0.09
    æŃ£å¥½
    -0.09
    å·²ä¹ħ
    -0.09
    æĪĸåħ¶ä»ĸ
    -0.08
    åĽ«
    -0.08
    æµij
    -0.08
    åijĹ
    -0.08
    èĬĬ
    -0.08
    POSITIVE LOGITS
    really
    0.10
    羣çļĦ
    0.09
     lately
    0.09
    éķĽ
    0.08
     Truly
    0.08
    æ·±åħ¥
    0.08
    羣çļĦæĺ¯
    0.08
     guts
    0.08
     Really
    0.08
    å½»åºķ
    0.08
    Activations Density 0.122%

    No Known Activations