© 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. 6766
    Prev
    Next
    INDEX
    Explanations

    doom, death, demise, grave, ruin, fate

    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
    æ··
    -0.10
    æ±Ĺ
    -0.10
    ç½ķ
    -0.09
    zilla
    -0.09
    erval
    -0.08
    zu
    -0.08
    俨
    -0.08
    ragon
    -0.08
    primitive
    -0.08
    æĬĸ
    -0.08
    POSITIVE LOGITS
    æŃ»äº¡
    0.18
     death
    0.17
     demise
    0.17
    death
    0.16
    -death
    0.15
    çģŃ亡
    0.15
    çļĦåij½è¿IJ
    0.14
    身亡
    0.14
    æŃ»äºĨ
    0.14
    Death
    0.13
    Activations Density 0.054%

    No Known Activations