© 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. 55617
    Prev
    Next
    INDEX
    Explanations

    nuclear, radioactive, waste, disaster, emergency

    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
    TTY
    -0.11
     Reform
    -0.11
    æĶ»
    -0.10
     миÑĢов
    -0.09
    cé
    -0.09
    泪
    -0.09
    cott
    -0.09
    rice
    -0.09
    秸
    -0.09
    è¿°
    -0.09
    POSITIVE LOGITS
    éĴĬ
    0.12
    æł¸
    0.12
    æł¸çĶµ
    0.11
     nuclear
    0.11
    éĺ²æĬ¤
    0.10
    æĶ¾å°Ħ
    0.10
    éģ¿
    0.09
    éĵĢ
    0.09
     radioactive
    0.09
    çħ§å°Ħ
    0.09
    Activations Density 0.135%

    No Known Activations