© 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. 62101
    Prev
    Next
    INDEX
    Explanations

    body parts

    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
     parity
    -0.09
    roys
    -0.09
    راÙħ
    -0.09
     PD
    -0.08
     ÑĢож
    -0.08
    çĿ¦
    -0.08
    ç¹ģèį£
    -0.08
    è¡ĮåĪĹ
    -0.08
    terra
    -0.08
    éĴ¤
    -0.08
    POSITIVE LOGITS
    éĥ¨ä½į
    0.24
    åĻ¨å®ĺ
    0.15
    躯
    0.15
    åħ¨èº«
    0.14
    Parts
    0.14
     Parts
    0.14
    éĥ¨ä»¶
    0.13
    äºĶå®ĺ
    0.13
     parts
    0.13
     regions
    0.12
    Activations Density 0.175%

    No Known Activations