© 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. 34536
    Prev
    Next
    INDEX
    Explanations

    numbers followed by specific descriptors

    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
    åĬłç´§
    -0.09
    emes
    -0.08
    sig
    -0.08
    /AP
    -0.08
    人身
    -0.08
    é¾ĻéŨ
    -0.08
     Obl
    -0.08
    é¦ĸåħĪè¦ģ
    -0.08
    (contents
    -0.08
    oon
    -0.07
    POSITIVE LOGITS
    è·ijäºĨ
    0.10
    åĥįãģį
    0.09
    onium
    0.09
    IDER
    0.09
    brook
    0.09
    ãģ£ãģ
    0.08
    æĪIJ份
    0.08
    ç»ĿæľĽ
    0.08
    ÑīиÑĤ
    0.08
    ç³ģ
    0.08
    Activations Density 0.110%

    No Known Activations