© 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. 23089
    Prev
    Next
    INDEX
    Explanations

    ending mischief

    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
    ush
    -0.09
    -ons
    -0.08
    å§Ķåī¯ä¹¦è®°
    -0.08
    姿åĬ¿
    -0.08
    ons
    -0.08
    itions
    -0.08
     (
    -0.08
    t
    -0.08
    ex
    -0.08
    Ln
    -0.08
    POSITIVE LOGITS
    eker
    0.10
    StringEncoding
    0.09
    ItemAt
    0.09
    éĤ£ä»½
    0.09
    çıŃ
    0.09
    ủng
    0.09
    elon
    0.09
    oba
    0.09
    >Date
    0.09
    åĭĥåĭĥ
    0.09
    Activations Density 0.001%

    No Known Activations