© 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. 29924
    Prev
    Next
    INDEX
    Explanations

    Feldmann, Feldman, Borgfeld, Grönefeld, Schönefeld

    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
    finder
    -0.10
    iating
    -0.09
    éĺ³åħī
    -0.09
    羣çĪ±
    -0.08
    ãģĬå¾Ĺ
    -0.08
    æĮĤçĿĢ
    -0.08
    iates
    -0.08
     Greens
    -0.08
    finding
    -0.08
     spr
    -0.08
    POSITIVE LOGITS
    wi
    0.10
    uki
    0.09
    neo
    0.09
    igma
    0.09
    lli
    0.08
    .au
    0.08
    aurus
    0.08
    ito
    0.08
    mp
    0.08
     Rex
    0.08
    Activations Density 0.004%

    No Known Activations