© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36120
    Prev
    Next
    INDEX
    Explanations

    child porn and abuse

    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
    èĽ®
    -0.10
    .Lines
    -0.09
    nost
    -0.09
    åĽ¢ç»ĵ
    -0.08
    ìķ½
    -0.08
    乡
    -0.08
    (success
    -0.08
    建çŃijä¸ļ
    -0.08
    æŃ»åIJİ
    -0.08
    æĶĢ
    -0.08
    POSITIVE LOGITS
    iped
    0.10
    ãĤ³ãĥ³ãĥĨãĥ³ãĥĦ
    0.09
    å¤ļåªĴä½ĵ
    0.09
    åŃ©åŃIJçļĦ
    0.09
    anela
    0.09
    anje
    0.09
    é¡·
    0.09
    æĭ·
    0.08
    éľĩæħij
    0.08
    è¾ħ导åijĺ
    0.08
    Activations Density 0.164%

    No Known Activations