© 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. 10978
    Prev
    Next
    INDEX
    Explanations

    obviously introducing self-evident points

    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
    ãģĿãĤĤãģĿãĤĤ
    -0.10
    éªĬ
    -0.09
    çĽ´è¾¾
    -0.09
    колÑĮ
    -0.09
    åħ³éĶ®æĺ¯
    -0.09
    ä¸ĭ称
    -0.08
    inde
    -0.08
    ä¹ĥ
    -0.08
    ä¸Ģèĩ´æĢ§
    -0.08
    UES
    -0.08
    POSITIVE LOGITS
    mente
    0.12
     (?)
    0.10
    ity
    0.10
    erved
    0.10
    ãģªãģĮãĤī
    0.10
    æĻ°
    0.10
    -minded
    0.09
    ly
    0.09
    çĦ¶æĺ¯
    0.09
    vir
    0.09
    Activations Density 0.012%

    No Known Activations