© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9254
    Prev
    Next
    INDEX
    Explanations

    a bit of a

    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
    è´Łæĭħ
    -0.09
    áµ¥
    -0.09
    å·´é»İ
    -0.08
     ÙĨÙĤ
    -0.08
    anke
    -0.08
    oon
    -0.08
    ä¸ĢçĤ¹çĤ¹
    -0.08
    éĥ¨åĪĨåĨħ容
    -0.08
    ONSE
    -0.08
    cke
    -0.08
    POSITIVE LOGITS
    èĩ³æѤ
    0.10
     Redistributions
    0.10
    enos
    0.09
    åĪĽä¸ĭ
    0.09
    zelf
    0.09
    å°±è¿Ļæł·
    0.09
    è°ľ
    0.09
    lox
    0.09
    mpp
    0.09
    lich
    0.09
    Activations Density 0.004%

    No Known Activations