© 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. 4555
    Prev
    Next
    INDEX
    Explanations

    simple

    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
    éľĦ
    -0.10
    åľľ
    -0.09
    åħ·ä½ĵæĥħåĨµ
    -0.09
    å°ıå¿ĥ翼翼
    -0.09
    好äºĨ
    -0.09
    奥æŀĹ
    -0.09
    èĥ¼
    -0.08
    ä»ħä¾Ľ
    -0.08
    è§Ī
    -0.08
    åIJĦç§į
    -0.08
    POSITIVE LOGITS
    /simple
    0.21
    ç®ĢåįķçļĦ
    0.21
    ç®Ģåįķ
    0.19
    .Simple
    0.19
     simple
    0.16
     Simple
    0.15
     SIMPLE
    0.15
    éĤ£ä¹Īç®Ģåįķ
    0.15
     simples
    0.15
    ç°¡åĸ®
    0.15
    Activations Density 0.190%

    No Known Activations