© 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. 7556
    Prev
    Next
    INDEX
    Explanations

    hang

    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
    çĩİ
    -0.10
    èĪĮå°ĸ
    -0.09
    andom
    -0.08
    åİ»è¿ĩ
    -0.08
    æĪ´ä¸Ĭ
    -0.08
    éĻĤ
    -0.08
    oret
    -0.08
    ä¸ĢéĹ´
    -0.08
    obar
    -0.08
    pest
    -0.08
    POSITIVE LOGITS
    rah
    0.09
    æĿŁ
    0.09
    undle
    0.09
    象
    0.09
    ä¿Ĺ
    0.09
    æĬµæĬ¼
    0.09
    GBK
    0.09
    éĤ®å¯Ħ
    0.09
    侥幸
    0.08
     Manus
    0.08
    Activations Density 0.004%

    No Known Activations