© 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. 39653
    Prev
    Next
    INDEX
    Explanations

    image file formats JPEG MPEG

    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
    peak
    -0.08
    æľīæĿĥ
    -0.08
    çıī
    -0.08
    åĭĥåĭĥ
    -0.08
     пиÑĤ
    -0.08
    ðŁı»
    -0.08
    çŁŃæŶéĹ´åĨħ
    -0.08
    BY
    -0.08
    spb
    -0.08
    æľīè¿ĩ
    -0.08
    POSITIVE LOGITS
    ../../../
    0.11
     dro
    0.09
    edy
    0.09
    æĬĵä½ı
    0.09
    uest
    0.09
    idos
    0.09
    letes
    0.09
    大åݦ
    0.08
    æ±ģ
    0.08
    âĺ°
    0.08
    Activations Density 0.012%

    No Known Activations