© 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. 32087
    Prev
    Next
    INDEX
    Explanations

    by

    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
    è¯ī
    -0.10
    NaN
    -0.09
    æĪ¿è½¦
    -0.09
    ä¸ĢéĿ¢
    -0.09
    еÑģÑĤв
    -0.09
    æĭĽ
    -0.09
    é«ĺéĵģ
    -0.09
     наг
    -0.08
    åĿł
    -0.08
    æĵ¦
    -0.08
    POSITIVE LOGITS
    vol
    0.10
    .bs
    0.09
    cka
    0.09
    íĮIJ
    0.08
    stroy
    0.08
    _HAL
    0.08
    Kn
    0.08
     пÑĢаво
    0.08
    è¤IJ
    0.08
    æĹłæĥħ
    0.08
    Activations Density 0.088%

    No Known Activations