© 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. 18874
    Prev
    Next
    INDEX
    Explanations

    Pharrell Williams, Bofors gun, Dbpedia

    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
    omi
    -0.10
    trinsic
    -0.10
    opal
    -0.10
    åĩŃ
    -0.10
    æĬĬæĪij
    -0.09
    esk
    -0.08
    è¡·
    -0.08
    adden
    -0.08
    ptest
    -0.08
    å½ķ
    -0.08
    POSITIVE LOGITS
    untu
    0.10
    åĬ¨åijĺ
    0.09
    æĦıè§ģåıįé¦Ī
    0.09
    èĥĻ
    0.08
     Rich
    0.08
    åIJī祥
    0.08
     dow
    0.08
    antes
    0.08
    enal
    0.08
    onian
    0.08
    Activations Density 0.007%

    No Known Activations