© 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. 28007
    Prev
    Next
    INDEX
    Explanations

    i thought i would

    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
    æ·ĸ
    -0.10
    ople
    -0.09
    åį±éĻ©
    -0.09
     hint
    -0.08
    Pt
    -0.08
    æĺ¯åĽ½åĨħ
    -0.08
    hra
    -0.08
     competent
    -0.08
    æĬij
    -0.08
    åħħæ²Ľ
    -0.08
    POSITIVE LOGITS
    çŃł
    0.12
    ãĤ·ãĤ§
    0.10
    æĥ³åĪ°äºĨ
    0.10
    è¶ģ
    0.10
    åĮ¹
    0.10
    äºĨè¿ĩæĿ¥
    0.09
    ç»ĻæĤ¨
    0.09
    æĿ¥çľĭçľĭ
    0.09
    æĥ³åĪ°
    0.09
    çľĭäºĨçľĭ
    0.09
    Activations Density 0.169%

    No Known Activations