© 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. 62486
    Prev
    Next
    INDEX
    Explanations

    Trent and place names

    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
    äºijéĽĨ
    -0.10
    fare
    -0.10
    容
    -0.10
    æĪ¢
    -0.09
    ffield
    -0.09
    æĩĤå¾Ĺ
    -0.09
    èĥĻ
    -0.09
    appid
    -0.09
    å¼¢
    -0.09
    ksam
    -0.09
    POSITIVE LOGITS
    matic
    0.09
    éĻ¤æѤ
    0.09
    æĿľåħ°
    0.09
    ioned
    0.08
    åĴļ
    0.08
    æ°²
    0.08
    PCS
    0.08
    licated
    0.08
     aspir
    0.08
     Dover
    0.08
    Activations Density 0.008%

    No Known Activations