© 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. 22735
    Prev
    Next
    INDEX
    Explanations

    connects or links

    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
    èĥĮä¸Ĭ
    -0.10
    çļĦ人çĶŁ
    -0.09
     Walls
    -0.09
    ãĤĥ
    -0.08
    积
    -0.08
    缮çļĦåľ°
    -0.08
    å°±è§īå¾Ĺ
    -0.08
    å¹½
    -0.08
    身é«ĺ
    -0.08
    ramento
    -0.08
    POSITIVE LOGITS
    纽带
    0.19
    è´¯éĢļ
    0.17
    贯穿
    0.16
    æīĵéĢļ
    0.14
    è´¯
    0.14
     link
    0.13
    è¿ŀæİ¥
    0.13
    (connect
    0.12
     connect
    0.12
     connecting
    0.12
    Activations Density 0.172%

    No Known Activations