© 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. 30239
    Prev
    Next
    INDEX
    Explanations

    special characters and syntax markers

    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
    ä¸įäºĨ
    -0.08
    istic
    -0.08
    s
    -0.08
     fucking
    -0.08
    t
    -0.08
     various
    -0.07
    å¿ħå¤ĩ
    -0.07
     ar
    -0.07
    è¿Ŀ
    -0.07
    è¿ĺåı¯
    -0.07
    POSITIVE LOGITS
    atre
    0.11
    åĽŀçŃĶ
    0.11
    woord
    0.10
    Ø£Ùĩد
    0.10
    adero
    0.10
    erah
    0.10
    Slinky
    0.10
    oden
    0.10
    odore
    0.10
    ðŁĵĮ
    0.10
    Activations Density 0.006%

    No Known Activations