© 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. 48751
    Prev
    Next
    INDEX
    Explanations

    look for or at

    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
    带走
    -0.11
    赤
    -0.09
    ismo
    -0.09
    ä¹ĭå¿ĥ
    -0.08
     corpus
    -0.08
    åħīè¾ī
    -0.08
     Cust
    -0.08
    jsc
    -0.08
    irut
    -0.08
     NotImplementedError
    -0.08
    POSITIVE LOGITS
    æľĽåİ»
    0.11
    å¾Ģä¸Ĭ
    0.10
    æĬ¬å¤´
    0.09
    -spe
    0.09
    çľĭè§ģ
    0.09
    åij¨åĽ´
    0.09
    åij¨åĽ´çļĦ
    0.09
    èģĶè°Ĭ
    0.09
     Saw
    0.09
    à¸Ĺะ
    0.09
    Activations Density 0.027%

    No Known Activations