© 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. 54771
    Prev
    Next
    INDEX
    Explanations

    IEEE standards and academic references

    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
     limp
    -0.09
    gar
    -0.09
    伺
    -0.09
    abella
    -0.08
    è¾ij
    -0.08
    .tm
    -0.08
    解æĶ¾
    -0.08
     dur
    -0.08
    Äł
    -0.08
     Dur
    -0.08
    POSITIVE LOGITS
     Fatal
    0.10
    éªĦåĤ²
    0.09
    çļĦæĢģ度
    0.09
    Fatal
    0.09
    éļ¼
    0.09
    .bias
    0.09
     fatally
    0.09
    UCH
    0.09
     piss
    0.09
    çļĦåį°è±¡
    0.09
    Activations Density 0.027%

    No Known Activations