© 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. 18434
    Prev
    Next
    INDEX
    Explanations

    latex, marks, files

    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
     relaxed
    -0.09
    们
    -0.08
    Sexy
    -0.08
    fort
    -0.08
    ARS
    -0.08
    LOCKS
    -0.08
    gap
    -0.08
    onical
    -0.08
     Tin
    -0.07
    AGR
    -0.07
    POSITIVE LOGITS
    çĹķ
    0.10
    é£ŀæī¬
    0.10
    çłĶ究éĻ¢
    0.10
    çĹķ迹
    0.10
    ç¥ĸåĽ½
    0.10
    aylor
    0.09
    erus
    0.09
    éĢıè¿ĩ
    0.09
     desn
    0.09
    顺çķħ
    0.09
    Activations Density 0.005%

    No Known Activations