© 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. 11823
    Prev
    Next
    INDEX
    Explanations

    forwarded 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
    ening
    -0.13
    ness
    -0.10
    ings
    -0.10
    heet
    -0.10
    ened
    -0.10
    ðŁı»
    -0.09
    ï¸ı
    -0.09
    ee
    -0.09
    hips
    -0.09
    ments
    -0.09
    POSITIVE LOGITS
    èĻ±
    0.09
    itti
    0.08
    ä¸İåħ¶
    0.08
    åĹĿ
    0.08
    inqu
    0.08
    åĪ«è¯´
    0.08
    Ĥæķ°
    0.08
    æĥ³çŁ¥éģĵ
    0.08
     Bahamas
    0.08
    Haz
    0.08
    Activations Density 0.273%

    No Known Activations