© 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. 57199
    Prev
    Next
    INDEX
    Explanations

    Microsoft Office file formats

    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
    带æľī
    -0.08
     Chap
    -0.08
    auer
    -0.08
    entifier
    -0.08
    æĦŁåĬ¨
    -0.08
    omnia
    -0.08
     Plat
    -0.08
    åIJ¦
    -0.08
     CMP
    -0.08
    retval
    -0.07
    POSITIVE LOGITS
    дол
    0.11
    大使
    0.10
    寿
    0.10
    çĿ
    0.09
    æıŃ
    0.09
    绥
    0.09
    驯
    0.09
    被æįķ
    0.09
    縻
    0.09
    箫
    0.09
    Activations Density 0.205%

    No Known Activations