© 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. 27792
    Prev
    Next
    INDEX
    Explanations

    S followed by specific words

    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
     Form
    -0.10
    ole
    -0.09
    æĻ°
    -0.09
    çĶŁç´ł
    -0.08
    hole
    -0.08
    ifar
    -0.08
    Form
    -0.08
    man
    -0.08
    sell
    -0.08
    heimer
    -0.08
    POSITIVE LOGITS
    ãĥĬãĥ«
    0.10
    以èī²
    0.10
    åħĨ
    0.09
    æľºåŀĭ
    0.09
    çĨ
    0.09
    __$
    0.09
    寤
    0.09
    elen
    0.09
    ä½Ļ
    0.09
    æĤ£æľī
    0.08
    Activations Density 0.017%

    No Known Activations