© 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. 18029
    Prev
    Next
    INDEX
    Explanations

    death and doom

    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
    对ä¸įèµ·
    -0.10
    æīª
    -0.09
    äºĴè¡¥
    -0.09
    åıijæī¬
    -0.08
    èµŀèµı
    -0.08
    åħ¶éĹ´
    -0.08
    èĶ¼
    -0.08
    ĪæĿĥ
    -0.08
    ç»·
    -0.08
    èĥĮçĿĢ
    -0.08
    POSITIVE LOGITS
    æŃ»äº¡
    0.20
     death
    0.18
    death
    0.16
     doom
    0.15
    çĶŁåij½çļĦ
    0.15
    çĶŁæŃ»
    0.14
    -death
    0.13
    _death
    0.13
     Death
    0.12
    æŃ»åĪij
    0.12
    Activations Density 0.080%

    No Known Activations