© 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. 27720
    Prev
    Next
    INDEX
    Explanations

    top or shirt

    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
    åĽ´
    -0.10
     Bod
    -0.09
     unh
    -0.09
    esar
    -0.08
    Sink
    -0.08
     Lew
    -0.08
    rec
    -0.08
    æĴŃ
    -0.08
     unto
    -0.08
    esse
    -0.08
    POSITIVE LOGITS
    çĥŃå¿ĥ
    0.09
    peed
    0.09
    æĬĵä½ı
    0.09
     Dias
    0.09
     FAC
    0.09
    æ´½
    0.09
    innen
    0.09
    ẫ
    0.08
    .SetActive
    0.08
    oret
    0.08
    Activations Density 0.003%

    No Known Activations