© 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. 49259
    Prev
    Next
    INDEX
    Explanations

    lack of

    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
    manuel
    -0.10
    ior
    -0.10
    alf
    -0.09
    ctions
    -0.09
    ä¸Ģå°ģ
    -0.09
    eries
    -0.09
    ANDLE
    -0.09
    iro
    -0.09
    iore
    -0.09
    íľ´
    -0.08
    POSITIVE LOGITS
    lessly
    0.11
    ened
    0.11
    ening
    0.10
    ä¹ı
    0.10
    uyá»ģn
    0.09
    ively
    0.09
    å®īåħ¨æĦŁ
    0.09
    alah
    0.09
    ол
    0.09
    æĸĹå¿Ĺ
    0.09
    Activations Density 0.016%

    No Known Activations