© 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. 35351
    Prev
    Next
    INDEX
    Explanations

    quotation mark followed by sentence start

    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
    es
    -0.10
    ëį°
    -0.10
    ed
    -0.09
    enticate
    -0.09
    enticated
    -0.09
    ers
    -0.09
    eling
    -0.09
    æĥħèĬĤ
    -0.09
    al
    -0.08
    .uk
    -0.08
    POSITIVE LOGITS
    åIJĦèĩªçļĦ
    0.10
    ees
    0.10
    pane
    0.09
    åİ¢
    0.09
     Oro
    0.09
    udad
    0.09
    çľģæĶ¿åºľ
    0.09
    ãģijãģ¦
    0.08
     dise
    0.08
    legate
    0.08
    Activations Density 0.066%

    No Known Activations