© 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. 4760
    Prev
    Next
    INDEX
    Explanations

    repurpose, recycle, value, potential

    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
    æĦŁåıĹ
    -0.09
     vere
    -0.09
    å¹³è¡Į
    -0.09
    ÂŃs
    -0.09
    æĭ¼éŁ³
    -0.09
     rider
    -0.08
    ç¬¬åĽĽå±Ĭ
    -0.08
    çĿĢæĢ¥
    -0.08
    lon
    -0.08
    éĽªå±±
    -0.08
    POSITIVE LOGITS
    æŧ
    0.13
    综åIJĪåĪ©çĶ¨
    0.13
    转åĮĸ
    0.11
    æĶ¹éĢł
    0.11
    èĪĬ
    0.10
    转åĮĸ为
    0.10
    é¾Ħ
    0.10
    æ®ĭ
    0.10
    转æį¢
    0.10
    æĪij以为
    0.09
    Activations Density 0.065%

    No Known Activations