© 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. 17487
    Prev
    Next
    INDEX
    Explanations

    damage, death, or health

    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
    驾é©Ń
    -0.10
    串èģĶ
    -0.10
    èĴľ
    -0.09
    AVE
    -0.09
    念
    -0.09
    æºľ
    -0.09
    ĥģ
    -0.09
    éĵ²
    -0.09
    ç§Ģ丽
    -0.09
    é»İ
    -0.09
    POSITIVE LOGITS
    伤亡
    0.11
    è¡Ģ
    0.11
    åıĹ伤
    0.11
    身亡
    0.10
    é²ľè¡Ģ
    0.10
    çĮ®è¡Ģ
    0.10
    çĶŁåij½çļĦ
    0.10
    æŃ»äºĨ
    0.10
    失
    0.10
     blood
    0.09
    Activations Density 0.084%

    No Known Activations