© 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. 22402
    Prev
    Next
    INDEX
    Explanations

    write or reading/writing

    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
    éĶ¯
    -0.10
    odash
    -0.09
     Fir
    -0.09
    /Area
    -0.09
    å½ĵåľº
    -0.09
    .kotlin
    -0.09
    ä¾Ŀæ³ķ追究
    -0.09
    è§ģè¯ģäºĨ
    -0.09
    å«Į
    -0.09
    -redux
    -0.09
    POSITIVE LOGITS
    ble
    0.11
    äºĨä¸įå°ij
    0.10
     onto
    0.10
    æĹłçŁ¥
    0.10
    enstein
    0.09
    饱满
    0.09
    /read
    0.09
    äºĨ许å¤ļ
    0.09
    utes
    0.09
    /write
    0.08
    Activations Density 0.178%

    No Known Activations