© 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. 10748
    Prev
    Next
    INDEX
    Explanations

    declared, deemed, judged, or itself

    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
    æıĨ
    -0.12
    ä¸įæĪIJ
    -0.11
     indef
    -0.11
    çļĦåĪ©çĽĬ
    -0.10
    ìķ
    -0.09
    æ³ķå¾ĭè§Ħå®ļ
    -0.09
    etti
    -0.09
    edar
    -0.09
    vik
    -0.09
    vey
    -0.09
    POSITIVE LOGITS
    è¿ĺ好
    0.10
     part
    0.09
     unable
    0.09
     able
    0.08
    æľĢç¾İ
    0.08
    ä¸Ģä½ĵ
    0.08
    ng
    0.08
    å¤ļ
    0.08
    驹
    0.08
     alive
    0.08
    Activations Density 0.070%

    No Known Activations