© 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. 2134
    Prev
    Next
    INDEX
    Explanations

    meadows and fields

    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
    ota
    -0.10
    çģ¶
    -0.10
    è¿Ľé£Ł
    -0.09
    procs
    -0.09
    æİĮ
    -0.09
    称ä¹ĭ
    -0.09
    家å±ŀ
    -0.09
    annel
    -0.09
    Drawable
    -0.09
    ye
    -0.09
    POSITIVE LOGITS
    ÙĦاÙĨ
    0.10
    edException
    0.09
    croft
    0.09
    oodle
    0.08
    ร
    0.08
    lined
    0.08
    éĩĮçļĦ
    0.08
    ing
    0.08
    anken
    0.08
    ed
    0.08
    Activations Density 0.020%

    No Known Activations