© 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. 5123
    Prev
    Next
    INDEX
    Explanations

    low, high, flat, drained, elevated descriptions

    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
    à¹Ĥย
    -0.10
    ฤ
    -0.09
    olith
    -0.09
    èĬŁ
    -0.09
    afari
    -0.09
    å¥ĸåĬ±
    -0.09
    éºŁ
    -0.09
    itest
    -0.09
     veto
    -0.09
    æīĵæĬĺ
    -0.09
    POSITIVE LOGITS
    洪水
    0.10
    åľºåľ°
    0.09
    æİĴæ°´
    0.09
    ä¼ĺè¶Ĭ
    0.08
    æµģåĬ¨
    0.08
    积水
    0.08
    åĢº
    0.08
    ä¸Ģçīĩ
    0.08
    ÑĥеÑĤ
    0.08
    wor
    0.08
    Activations Density 0.031%

    No Known Activations