© 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. 5102
    Prev
    Next
    INDEX
    Explanations

    keeping things limited

    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
    Berry
    -0.09
    oub
    -0.09
    åħ¸
    -0.08
    quate
    -0.08
    éĺ
    -0.08
    ä½ķåĨµ
    -0.08
    ocker
    -0.08
    éĢIJæ¸IJ
    -0.08
    å¡Į
    -0.08
    entine
    -0.08
    POSITIVE LOGITS
    å°½éĩı
    0.11
    ä¸įè¶ħè¿ĩ
    0.10
    以åħį
    0.09
    éĻIJ度
    0.09
    ç®Ģåįķ
    0.08
    5
    0.08
    çķ´
    0.08
    2
    0.08
     Colum
    0.08
    åĩ¤
    0.08
    Activations Density 0.029%

    No Known Activations