© 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. 23788
    Prev
    Next
    INDEX
    Explanations

    specifying quality of group or type

    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
    å¦ĤæľŁ
    -0.08
    次ä¼ļè®®
    -0.08
    Alias
    -0.08
    åĦĭ
    -0.08
    好转
    -0.08
    èİŀ
    -0.08
    èĭ·
    -0.08
    acle
    -0.08
    nton
    -0.08
    éĽī
    -0.08
    POSITIVE LOGITS
    æĭĽçīĮ
    0.11
    æŁĵ
    0.11
    å§ĵ
    0.10
    asher
    0.10
    	TokenName
    0.10
    åħ¨æĺ¯
    0.10
    GREE
    0.09
    PAL
    0.09
    quarters
    0.09
    .non
    0.09
    Activations Density 0.242%

    No Known Activations