© 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. 54494
    Prev
    Next
    INDEX
    Explanations

    decision or mistake

    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
    éŀł
    -0.09
    éĩįéĩį
    -0.09
    综åIJĪæ²»çIJĨ
    -0.09
    趯
    -0.09
    .pref
    -0.09
    uff
    -0.09
    ç¼´
    -0.09
    enal
    -0.08
    æĢł
    -0.08
    Leaks
    -0.08
    POSITIVE LOGITS
    åĨ³çŃĸ
    0.15
    çļĦéĢīæĭ©
    0.13
     decision
    0.13
    æĺİæĻº
    0.13
    Decision
    0.12
    éĶĻäºĨ
    0.11
     Decision
    0.11
    decision
    0.11
     choice
    0.11
     decisión
    0.11
    Activations Density 0.038%

    No Known Activations