© 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. 17113
    Prev
    Next
    INDEX
    Explanations

    cut through outline

    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
    å½ķ
    -0.12
    èįĥ
    -0.11
    å²Ĺ
    -0.10
    oop
    -0.10
    .URI
    -0.10
    UGINS
    -0.09
    æ¤į
    -0.09
    .Safe
    -0.09
     FAG
    -0.09
    åħ¥å¢ĥ
    -0.09
    POSITIVE LOGITS
    ä¸Ģå±Ĥ
    0.10
    åĨ²
    0.10
    ä¸Ģéģĵ
    0.10
    ä¸Ģçľĭ
    0.09
    ä¸Ģçľ¼
    0.08
    ather
    0.08
    éĨį
    0.08
    è¶Ĭ
    0.08
    æ¾Ħæ¸ħ
    0.08
    ä¸ĢéĹ´
    0.08
    Activations Density 0.032%

    No Known Activations