© 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. 58928
    Prev
    Next
    INDEX
    Explanations

    too good to be true

    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
    до
    -0.10
    æ³Ľ
    -0.08
    Interested
    -0.08
    OME
    -0.08
    APT
    -0.08
    竦
    -0.08
    çľĭä¼¼
    -0.08
    æĪij对
    -0.08
    æŃĮ
    -0.08
     Pike
    -0.08
    POSITIVE LOGITS
     quot
    0.09
     Haram
    0.09
    çħŀ
    0.09
    æIJģ
    0.09
    诬
    0.09
    otes
    0.08
    对å¤ĸå¼ĢæĶ¾
    0.08
    uze
    0.08
    è¯Ĩ
    0.08
    å°ıç»Ħ
    0.08
    Activations Density 0.009%

    No Known Activations