© 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. 61914
    Prev
    Next
    INDEX
    Explanations

    illegal activities and items

    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
    ableView
    -0.11
    manship
    -0.10
    abant
    -0.10
    ãĥ³ãĤ°
    -0.09
    iller
    -0.09
    ingly
    -0.09
    bull
    -0.09
    çıª
    -0.09
    combe
    -0.09
    大å±Ģ
    -0.09
    POSITIVE LOGITS
    ities
    0.17
    ITY
    0.14
    /il
    0.12
    ity
    0.11
    mente
    0.11
    -minded
    0.10
    itas
    0.10
    ism
    0.10
    ITIES
    0.10
    å¾Ĵ
    0.10
    Activations Density 0.008%

    No Known Activations