© 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. 58750
    Prev
    Next
    INDEX
    Explanations

    communication technologies and fields

    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
    æĺ¯éŁ©åĽ½
    -0.11
     Korean
    -0.10
    éŁ©åĽ½
    -0.09
     Japanese
    -0.08
     Korea
    -0.08
     Chinese
    -0.08
    AREN
    -0.08
     Friendship
    -0.08
    éĢļ讯
    -0.08
     Duty
    -0.08
    POSITIVE LOGITS
    éĤij
    0.10
    ÑģеÑĢ
    0.10
    éű
    0.09
    èİ·
    0.09
    éŁŃ
    0.09
    æ±°
    0.09
     Narc
    0.09
    è§Ħ
    0.09
    æľªæĪIJ
    0.09
    elize
    0.09
    Activations Density 1.099%

    No Known Activations