© 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. 26642
    Prev
    Next
    INDEX
    Explanations

    Sherlock Holmes detective

    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
    enko
    -0.11
    ester
    -0.10
    åħľ
    -0.09
    éŃģ
    -0.09
    pons
    -0.09
    .Std
    -0.09
    bbing
    -0.09
    .promise
    -0.09
    лиÑĨ
    -0.09
    ibel
    -0.09
    POSITIVE LOGITS
    mania
    0.10
    rium
    0.09
    ä¸ĭåįĬ
    0.09
    ian
    0.08
     haystack
    0.08
     Holmes
    0.08
    æݨçIJĨ
    0.08
     hans
    0.08
    æ°
    0.08
    sson
    0.08
    Activations Density 0.013%

    No Known Activations