© 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. 35035
    Prev
    Next
    INDEX
    Explanations

    kid inside

    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
    ing
    -0.11
    eda
    -0.10
    èĤ¢
    -0.09
     rhythm
    -0.09
    eron
    -0.09
    ignal
    -0.08
    ï¸ı
    -0.08
    ÑģÑĤÑĥп
    -0.08
    深深çļĦ
    -0.08
    ãĤĪãģĨãģ«
    -0.08
    POSITIVE LOGITS
    ress
    0.11
    awan
    0.10
    æĢ»å±Ģ
    0.10
    neys
    0.10
    owski
    0.09
    åĢij
    0.09
    amic
    0.09
    roc
    0.09
    rary
    0.09
    éĻĭ
    0.09
    Activations Density 0.008%

    No Known Activations