© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10386
    Prev
    Next
    INDEX
    Explanations

    rules and logic

    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
    rip
    -0.09
    æĹĭ
    -0.08
    Streamer
    -0.08
    åħ¨ä½ĵ
    -0.08
    OPS
    -0.08
    社ä¼ļç»Ħç»ĩ
    -0.08
    åŀĮ
    -0.08
    çľĭä¸įè§ģ
    -0.07
    åĽ½éĻħåIJĪä½ľ
    -0.07
     pill
    -0.07
    POSITIVE LOGITS
    /rules
    0.12
    è§ĦåĪĻ
    0.11
    çŁ¥è¯Ĩ
    0.10
     rules
    0.10
    RuleContext
    0.10
     ÑįкÑģпеÑĢÑĤ
    0.10
    .Rule
    0.10
    .games
    0.09
    æݨçIJĨ
    0.09
    rules
    0.09
    Activations Density 0.086%

    No Known Activations