© 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. 63372
    Prev
    Next
    INDEX
    Explanations

    real bad

    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
    æĿŀ
    -0.09
    ride
    -0.09
    糸
    -0.09
     Abyss
    -0.08
    AMI
    -0.08
    è¯ķçĤ¹
    -0.08
    çݯèĬĤ
    -0.08
    èĪŀåı°ä¸Ĭ
    -0.08
    iture
    -0.08
    æģį
    -0.08
    POSITIVE LOGITS
    羣å®ŀçļĦ
    0.10
     IMPORTANT
    0.10
    羣å®ŀ
    0.10
    nar
    0.09
    çļĦéĤ£ç§į
    0.09
    ignment
    0.08
    .Formatter
    0.08
    ãģ§ãģĻãĤĪ
    0.08
     Bras
    0.08
    è°Ľ
    0.08
    Activations Density 0.017%

    No Known Activations