© 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. 33117
    Prev
    Next
    INDEX
    Explanations

    percentage or rate

    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
    åĨįå¤ļ
    -0.10
    urge
    -0.10
    é²ģ
    -0.09
     Redistribution
    -0.09
    ç»ķ
    -0.08
    èĩ
    -0.08
    _UNITS
    -0.08
     Limits
    -0.08
    èĤ©èĨĢ
    -0.08
    æĹłå¿§
    -0.08
    POSITIVE LOGITS
    çİĩ
    0.11
    ìľ¨
    0.10
    çİĩé¢Ĩ
    0.10
     rate
    0.09
    精确
    0.09
    filt
    0.09
     ratio
    0.09
    ratio
    0.09
    irate
    0.09
    åij½
    0.09
    Activations Density 0.027%

    No Known Activations