© 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. 62713
    Prev
    Next
    INDEX
    Explanations

    SB or sb followed by letters or punctuation

    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
    æĺ¯éĿŀ
    -0.12
    ENTS
    -0.09
     modest
    -0.09
    ertos
    -0.09
    iland
    -0.09
    ILLA
    -0.09
    ilee
    -0.08
    ed
    -0.08
    oz
    -0.08
    asn
    -0.08
    POSITIVE LOGITS
    æĥ³æĥ³
    0.10
    æĢ»åĨłåĨĽ
    0.10
    åħ¶éĹ´
    0.09
    arel
    0.09
    onian
    0.09
    缺åı£
    0.09
    é»į
    0.09
    mun
    0.08
    è£ħæī®
    0.08
    aking
    0.08
    Activations Density 0.040%

    No Known Activations