© 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. 27483
    Prev
    Next
    INDEX
    Explanations

    idea or prospect

    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
    äºĨåIJ§
    -0.09
    éĢļçķħ
    -0.09
    åģļå¾Ĺ
    -0.09
    æĻ°
    -0.09
    éħī
    -0.09
    磼
    -0.08
    èįł
    -0.08
    åĮĻ
    -0.08
    lient
    -0.08
    çķħéĢļ
    -0.08
    POSITIVE LOGITS
     prospect
    0.17
     idea
    0.15
     possibility
    0.15
    æĥ³åĪ°
    0.15
    idea
    0.14
     notion
    0.14
     prospects
    0.13
     возможноÑģÑĤи
    0.13
     ÙģÙĥرة
    0.13
    æĥ³æ³ķ
    0.13
    Activations Density 0.064%

    No Known Activations