© 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. 54467
    Prev
    Next
    INDEX
    Explanations

    hum or um followed by specific endings

    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
    OrUpdate
    -0.15
    xic
    -0.10
    çĴŀ
    -0.09
    çĤİ
    -0.08
    åIJĮæŶè¿ĺ
    -0.08
    们çļĦ
    -0.08
    itel
    -0.08
    ÑģÑĤÑĥп
    -0.08
    ../../../
    -0.08
    åĴķ
    -0.08
    POSITIVE LOGITS
    istic
    0.16
    ëį°
    0.10
    髦
    0.09
    è±¹
    0.09
     WithEvents
    0.09
    ize
    0.09
    èĶ¼
    0.08
    ãģĨãģ¡ãģ«
    0.08
    olare
    0.08
    è¹Ĵ
    0.08
    Activations Density 0.028%

    No Known Activations