© 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. 18229
    Prev
    Next
    INDEX
    Explanations

    copyright headers

    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
    /from
    -0.10
    serrat
    -0.10
    çľ¼çļ®
    -0.10
    ses
    -0.09
    entifier
    -0.09
    -called
    -0.09
    åIJĦ项
    -0.09
    /her
    -0.08
    é¢
    -0.08
    ifiable
    -0.08
    POSITIVE LOGITS
    orem
    0.19
    etheless
    0.18
    odore
    0.18
    malink
    0.13
    ÑįÑĤомÑĥ
    0.12
    gunakan
    0.12
    oret
    0.12
    atre
    0.12
    ories
    0.12
    pués
    0.11
    Activations Density 3.789%

    No Known Activations