© 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. 2092
    Prev
    Next
    INDEX
    Explanations

    impaired or not

    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
    å¼ĥ
    -0.10
    iges
    -0.09
     pist
    -0.09
     ÑģобÑģÑĤвенно
    -0.09
    DTD
    -0.09
    æ£Ħ
    -0.09
     Medina
    -0.08
    è¿Łè¿Ł
    -0.08
     separ
    -0.08
    .Paths
    -0.08
    POSITIVE LOGITS
    éĤ£ä¸ªæŶåĢĻ
    0.11
    æĥħ绪
    0.10
    å¿ĥæĥħ
    0.10
    çĬ¶æĢģä¸ĭ
    0.10
    éĤ£æŶ
    0.09
    ç³Ĭæ¶Ĥ
    0.09
     Mood
    0.09
     Sloan
    0.09
    çĿĢæĢ¥
    0.08
     mood
    0.08
    Activations Density 0.118%

    No Known Activations