© 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. 57167
    Prev
    Next
    INDEX
    Explanations

    making the best effort

    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
    ä¹į
    -0.09
    .hl
    -0.09
    ç¾Ĭæ¯Ľ
    -0.09
    å°±é¤IJ
    -0.09
    ä¸ĢäºĽ
    -0.09
    æŁ³
    -0.09
    åĮĹ大
    -0.09
    æѧ
    -0.09
     McMahon
    -0.09
    çĽĽå®´
    -0.09
    POSITIVE LOGITS
    lined
    0.10
    ä¸į让
    0.10
    æį¢äºĨ
    0.09
    åı£å²¸
    0.09
    éģ¿åħį
    0.09
     Dah
    0.08
    缩çŁŃ
    0.08
    积累äºĨ
    0.08
    çīº
    0.08
    å¸ļ
    0.08
    Activations Density 0.006%

    No Known Activations