© 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. 3162
    Prev
    Next
    INDEX
    Explanations

    advice or recommendation requests

    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
    æĦ¿
    -0.09
    çŃ·åŃIJ
    -0.09
    cow
    -0.09
    ä¹ĭæĹħ
    -0.09
    åĽĽå¤§
    -0.09
    ä»»
    -0.09
    çIJ®
    -0.09
    çļĦåĬĽéĩı
    -0.09
    å½¹
    -0.09
    æĺ¯å¦Ĥä½ķ
    -0.09
    POSITIVE LOGITS
    è¾Ļ
    0.10
    plier
    0.09
    _FUNC
    0.09
     mote
    0.08
    éħį
    0.08
    rels
    0.08
    éħįä¸Ĭ
    0.08
    à¸Ĺาà¸ĩ
    0.08
    pery
    0.08
    åħĦ
    0.08
    Activations Density 0.081%

    No Known Activations