© 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. 22876
    Prev
    Next
    INDEX
    Explanations

    attribution of text

    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
    alyze
    -0.09
    æīĭä¸ŃçļĦ
    -0.08
    bul
    -0.08
    ention
    -0.08
    rieve
    -0.08
    è§Ĵèī²
    -0.08
    OfWeek
    -0.08
    ç¤¾åĽ¢
    -0.08
    å½¢å¼ı
    -0.08
    ä¸ĢéĿ¢
    -0.08
    POSITIVE LOGITS
    plier
    0.10
    ä½ľé£İ
    0.09
    spath
    0.09
    äºĽä»Ģä¹Ī
    0.09
    ï¼ļ↵
    0.09
    autiful
    0.09
    outu
    0.09
    æ·¡æ·¡
    0.08
    ï¼ļ
    0.08
    SPATH
    0.08
    Activations Density 0.036%

    No Known Activations