© 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. 54730
    Prev
    Next
    INDEX
    Explanations

    tend to

    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
    ä»ħ
    -0.09
    ç²¼
    -0.09
    omed
    -0.09
    ../../../
    -0.08
     essentially
    -0.08
    trad
    -0.08
     ven
    -0.08
    ä¸Ģæłĭ
    -0.08
    举æĬ¥çĶµè¯Ŀ
    -0.08
    æĹĸ
    -0.08
    POSITIVE LOGITS
    ities
    0.10
    å®Įç¾İçļĦ
    0.10
    éĦ£
    0.10
    ently
    0.09
    lessly
    0.09
    elcome
    0.09
    orary
    0.09
    ouble
    0.09
    noch
    0.09
    ency
    0.09
    Activations Density 0.011%

    No Known Activations