© 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. 11274
    Prev
    Next
    INDEX
    Explanations

    fit in smaller space

    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
    omed
    -0.09
    èĸª
    -0.09
    é»ĦåŁĶ
    -0.09
    éķ¿æľŁ
    -0.09
    æݨ
    -0.08
    .Inv
    -0.08
    глаÑģ
    -0.08
    æ¤Ĵ
    -0.08
    parm
    -0.08
    éķ·æľŁ
    -0.08
    POSITIVE LOGITS
    æµĵ缩
    0.17
    çŁŃçŁŃ
    0.14
    çŁŃæŶéĹ´åĨħ
    0.13
    ç´§åĩij
    0.13
    åİĭ缩
    0.12
    åľ¨ä¸Ģ个
    0.12
    æľīéĻIJ
    0.12
     compact
    0.11
     compressed
    0.10
    åĮĨåĮĨ
    0.10
    Activations Density 0.093%

    No Known Activations