© 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. 4415
    Prev
    Next
    INDEX
    Explanations

    detachable, removable

    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
    è¹²
    -0.10
    ä¹Łåıªèĥ½
    -0.10
    ä¼ŀ
    -0.09
    åĸĿäºĨ
    -0.09
    åºŁ
    -0.09
    æģĥ
    -0.09
    äºĨä¸Ģåľº
    -0.08
    -pills
    -0.08
    _dropout
    -0.08
    yme
    -0.08
    POSITIVE LOGITS
     оÑĤд
    0.11
    erable
    0.09
    mani
    0.09
    以便
    0.09
    对äºİæĪij们
    0.08
    aily
    0.08
     easily
    0.08
    ìĶ©
    0.08
    大çIJĨ
    0.08
    èģĺ请
    0.08
    Activations Density 0.051%

    No Known Activations