© 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. 31592
    Prev
    Next
    INDEX
    Explanations

    vague, fuzzy, blur, dim

    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´±
    -0.09
    ä¸įåIJĥ
    -0.09
    çĪĨåıij
    -0.09
    _mono
    -0.08
    iku
    -0.08
    (chip
    -0.08
    éķĩ
    -0.08
    ิà¸ģ
    -0.08
    è¯Ī
    -0.08
    çĥ¹
    -0.08
    POSITIVE LOGITS
    模ç³Ĭ
    0.24
    ä¸įæ¸ħ
    0.17
    æľ¦èĥ§
    0.15
    æľ¦
    0.12
     fuzzy
    0.12
    éļIJ约
    0.12
     Blur
    0.11
    blur
    0.11
     vague
    0.11
    ä¸įæ¸ħæ¥ļ
    0.11
    Activations Density 0.098%

    No Known Activations