© 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. 1938
    Prev
    Next
    INDEX
    Explanations

    citing or citable

    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
    æŀģåĬĽ
    -0.10
    .synthetic
    -0.09
    uder
    -0.09
    çĿĢä»ĸ
    -0.08
    -readable
    -0.08
    spoken
    -0.08
    ±
    -0.08
    loads
    -0.08
    侧éĿ¢
    -0.08
     Buckley
    -0.08
    POSITIVE LOGITS
    è¡«
    0.10
    té
    0.10
    aison
    0.09
    亵
    0.09
    aines
    0.09
    ables
    0.09
    榴
    0.08
    phia
    0.08
    ight
    0.08
    å¨Ł
    0.08
    Activations Density 0.028%

    No Known Activations