© 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. 31137
    Prev
    Next
    INDEX
    Explanations

    tunnel vision

    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¸įåħ·å¤ĩ
    -0.10
    èĤ¤èī²
    -0.09
    个æĢ§
    -0.09
    éĥ¨åī¯
    -0.09
    éĵĥ
    -0.09
    enance
    -0.09
    zell
    -0.09
    稼
    -0.09
    çİĭçīĮ
    -0.09
    åĩºåĬ¨
    -0.09
    POSITIVE LOGITS
    éģ¥
    0.11
    éĽ¾
    0.10
    çľĭæ¸ħ
    0.09
     tunnel
    0.09
    é»ijæļĹ
    0.09
    mare
    0.09
    åħīæĺİ
    0.09
    ighted
    0.09
    ÙĦÙĬÙħ
    0.08
     mù
    0.08
    Activations Density 0.026%

    No Known Activations