© 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. 61350
    Prev
    Next
    INDEX
    Explanations

    could have

    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
     Bere
    -0.09
    è¿Ļå°±
    -0.09
    æĩĤå¾Ĺ
    -0.08
    èĥį
    -0.08
    ENTS
    -0.08
     bere
    -0.08
    ç°Į
    -0.08
     Gel
    -0.08
    uncan
    -0.08
     gel
    -0.08
    POSITIVE LOGITS
    éļıæŶ
    0.13
    éļ¨æĻĤ
    0.11
    è½»æĺĵ
    0.11
    åı¯ä»¥éĢīæĭ©
    0.10
    æľ¬å½ĵ
    0.10
    çŃĿ
    0.10
    æŀķ头
    0.09
    é¦į
    0.09
    è°ĥ
    0.09
    çļĦéĢīæĭ©
    0.09
    Activations Density 0.058%

    No Known Activations