© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 29777
    Prev
    Next
    INDEX
    Explanations

    void space or gaps

    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
    _OVERRIDE
    -0.09
    peri
    -0.08
    svp
    -0.08
    触
    -0.08
    æ¿Ĵ临
    -0.08
    æľ«ç«¯
    -0.08
    .Math
    -0.08
    irit
    -0.08
    ycle
    -0.08
     erw
    -0.08
    POSITIVE LOGITS
    éļĻ
    0.16
    缺åı£
    0.14
     gaps
    0.14
     gap
    0.13
    空çĻ½
    0.13
    空
    0.12
     void
    0.12
     space
    0.12
    ç½ħ
    0.11
    space
    0.11
    Activations Density 0.182%

    No Known Activations