© 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. 4777
    Prev
    Next
    INDEX
    Explanations

    adverse reactions and toxicity

    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
    onica
    -0.08
    è±³
    -0.08
    åĨ¢
    -0.08
    ÑģÑĢ
    -0.08
     VE
    -0.08
    srv
    -0.08
    æĿ¿åĿĹ
    -0.08
    ve
    -0.08
    VE
    -0.07
    -break
    -0.07
    POSITIVE LOGITS
    åī¯ä½ľçĶ¨
    0.13
    çĸĹæķĪ
    0.12
    æ¯ĴæĢ§
    0.11
    åıĹ害
    0.11
    æ¯Ĵ
    0.10
    è¶Ł
    0.10
    æķĪæŀľ
    0.09
    едÑĮ
    0.09
     доÑģÑĤ
    0.09
     toxicity
    0.09
    Activations Density 0.230%

    No Known Activations