© 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. 22648
    Prev
    Next
    INDEX
    Explanations

    since last time

    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
    ä¸ĢæŶ
    -0.10
    åİŁæĿ¥
    -0.10
    æľĢç»Ī
    -0.09
    emble
    -0.09
    leasing
    -0.09
    æŦ
    -0.09
     Ult
    -0.08
    æĮĩåIJij
    -0.08
    /graph
    -0.08
    ä¸į足以
    -0.08
    POSITIVE LOGITS
    é«ĺè´¨éĩı
    0.10
    çīŁ
    0.10
    utility
    0.10
    è¤Ĵ
    0.10
    æķĻåŃ¦è´¨éĩı
    0.09
    åı£æ°´
    0.09
    è°Ľ
    0.09
    åħ¬çĶ¨
    0.09
     quality
    0.08
    头
    0.08
    Activations Density 0.017%

    No Known Activations