© 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. 63092
    Prev
    Next
    INDEX
    Explanations

    updating lists 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
    popular
    -0.09
     popularity
    -0.09
     popular
    -0.08
    tec
    -0.08
    æĭħ忧
    -0.08
     DAMAGES
    -0.08
    åīĸæŀIJ
    -0.08
    تØŃÙĪÙĦ
    -0.08
    å®īå¨ľ
    -0.08
     Ñĩем
    -0.07
    POSITIVE LOGITS
    åĩĨç¡®æĢ§
    0.14
     accuracy
    0.12
    æĽ´æĸ°
    0.12
     Accuracy
    0.12
    åĩĨç¡®
    0.12
    çºłéĶĻ
    0.12
     accur
    0.11
    accuracy
    0.11
    çºłæŃ£
    0.11
    ä¿®æŃ£
    0.10
    Activations Density 0.068%

    No Known Activations