© 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. 39326
    Prev
    Next
    INDEX
    Explanations

    Category lists, external links, references

    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
    å¹»
    -0.08
    å°¤åħ¶æĺ¯åľ¨
    -0.08
    club
    -0.08
    æ©ĩ
    -0.08
    æĪIJåIJį
    -0.08
    å¹ħ度
    -0.08
     cons
    -0.07
    çľģ
    -0.07
     alleg
    -0.07
    antine
    -0.07
    POSITIVE LOGITS
    AGED
    0.10
    AGES
    0.10
    arro
    0.10
    éķ¡
    0.09
    geme
    0.09
    uctor
    0.09
    æijĦ
    0.09
     Disclaimer
    0.09
    è´Ł
    0.09
    rades
    0.09
    Activations Density 0.095%

    No Known Activations