© 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. 4357
    Prev
    Next
    INDEX
    Explanations

    standardized, uniform, common way

    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
     ÑģовеÑĢ
    -0.10
    æµģ失
    -0.09
    ç¯ļ
    -0.09
    åİŁè°ħ
    -0.09
    æį§
    -0.09
    å±ģ
    -0.09
    æĢ»ç»ĵ
    -0.09
    åIJĬ
    -0.09
    à¸Ļำ
    -0.09
    _FAST
    -0.08
    POSITIVE LOGITS
    è§ĦèĮĥ
    0.16
    绣ä¸Ģ
    0.16
    æłĩåĩĨåĮĸ
    0.15
    ä¸Ģèĩ´
    0.15
     standardized
    0.13
    æłĩåĩĨ
    0.13
    è§ĦèĮĥåĮĸ
    0.13
     across
    0.12
     consistency
    0.12
     consistent
    0.12
    Activations Density 0.153%

    No Known Activations