© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8894
    Prev
    Next
    INDEX
    Explanations

    abuse or tolerate shit

    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
    _LEAVE
    -0.10
     Cran
    -0.09
    èĮĥçķ´
    -0.09
    æĸ¹åľĨ
    -0.09
    éĦ¢
    -0.09
    å´©
    -0.08
    çŁ¥æĥħ
    -0.08
    Sensitive
    -0.08
    åĢı
    -0.08
    stad
    -0.08
    POSITIVE LOGITS
    å¿įåıĹ
    0.21
    容å¿į
    0.16
    å¿į
    0.14
    æī¿åıĹ
    0.13
     toler
    0.12
     tolerate
    0.12
    èĢIJå¿ĥ
    0.11
     tolerance
    0.10
    èĢIJ
    0.10
    åħĭæľį
    0.10
    Activations Density 0.099%

    No Known Activations