© 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. 48734
    Prev
    Next
    INDEX
    Explanations

    FAQ links and information

    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
    elay
    -0.09
    iability
    -0.09
    éģ£
    -0.08
    èĴ´
    -0.08
    gang
    -0.08
     Wend
    -0.08
    çĩİ
    -0.08
     Rudd
    -0.08
     amo
    -0.07
    åģ¥
    -0.07
    POSITIVE LOGITS
    ulous
    0.11
    ãģĤãĤĭç¨ĭ度
    0.10
    åħ¶å®ĥéĹ®é¢ĺ
    0.10
    -worthy
    0.10
    ÑĢак
    0.10
     sá»ķ
    0.09
    sWith
    0.09
    è¨ĢãģĪãģ°
    0.09
     mann
    0.09
    æµ·è¿IJ
    0.09
    Activations Density 0.008%

    No Known Activations