© 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. 62253
    Prev
    Next
    INDEX
    Explanations

    important/good followed by work/times/people/group/lifting/movies/yards/memories

    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
    æĹĨ
    -0.10
    åĮłå¿ĥ
    -0.09
    ils
    -0.09
    çĩ§
    -0.09
    pressions
    -0.09
    è¯ļæĦı
    -0.08
    chi
    -0.08
    -ahead
    -0.08
    åıĬåħ¶
    -0.08
    Ñĩи
    -0.08
    POSITIVE LOGITS
     ones
    0.13
    次è¦ģ
    0.10
    æľĢéļ¾
    0.09
     stuff
    0.09
    æĽ´éĩįè¦ģçļĦ
    0.09
    éĩįè¦ģçļĦ
    0.09
    羣æŃ£çļĦ
    0.09
    æĪı
    0.09
     Ones
    0.08
    æĽ´éĩįè¦ģ
    0.08
    Activations Density 0.058%

    No Known Activations