© 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. 39494
    Prev
    Next
    INDEX
    Explanations

    useless, unnecessary, or essential

    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
    ç¾ŀ
    -0.11
    æīĵçł´äºĨ
    -0.10
    ACE
    -0.10
     ace
    -0.10
    èĩªå¦Ĥ
    -0.10
    亲
    -0.09
    знаком
    -0.09
    åħį
    -0.09
    èĿİ
    -0.09
     illegal
    -0.08
    POSITIVE LOGITS
    没æľīå¿ħè¦ģ
    0.16
    ä¸įå¿ħè¦ģçļĦ
    0.15
    没å¿ħè¦ģ
    0.14
    éĻ¤éĿŀ
    0.12
     unnecessary
    0.12
    ä½ķå¿ħ
    0.11
    åĨĹ
    0.11
     êµ
    0.10
    å¤ļä½ĻçļĦ
    0.10
    å¤ļä½Ļ
    0.10
    Activations Density 0.193%

    No Known Activations