© 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. 3484
    Prev
    Next
    INDEX
    Explanations

    perceiving senses

    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
    çĿĢæīĭ
    -0.09
    CEL
    -0.08
    Compact
    -0.08
    å´İ
    -0.08
    wow
    -0.08
     Surprise
    -0.08
     uncont
    -0.08
    æ¤Ł
    -0.08
     PH
    -0.08
    GU
    -0.08
    POSITIVE LOGITS
    åIJ¬è§ģ
    0.16
    åIJ¬åĪ°
    0.14
    çľĭè§ģ
    0.11
    èģ½åĪ°
    0.11
     heard
    0.10
    çŁ¥æĻĵ
    0.10
     hears
    0.10
    读æĩĤ
    0.10
    hear
    0.10
    åIJ¬å¾Ĺ
    0.10
    Activations Density 0.049%

    No Known Activations