© 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. 11284
    Prev
    Next
    INDEX
    Explanations

    Mann, Mannan, Mannheim, Braunschweig

    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
    çĦ¶æĺ¯
    -0.11
    dess
    -0.10
    arded
    -0.10
    ssid
    -0.10
    ocode
    -0.10
    edly
    -0.10
    ertype
    -0.10
    eki
    -0.09
    ãĤĥ
    -0.09
    glass
    -0.09
    POSITIVE LOGITS
    ification
    0.11
    itol
    0.10
    woods
    0.10
    ities
    0.10
    å©¿
    0.09
    æ±ģ
    0.09
    ../../../
    0.09
    ism
    0.09
    keit
    0.09
    ify
    0.09
    Activations Density 0.005%

    No Known Activations