© 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. 22459
    Prev
    Next
    INDEX
    Explanations

    breakdown and down

    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
    endtime
    -0.09
     nhiên
    -0.09
    ÙĨÙī
    -0.09
    è¿ĩåĪĨ
    -0.09
    æ³Ľ
    -0.09
    erture
    -0.09
    atility
    -0.08
    åĩ»
    -0.08
    eous
    -0.08
    ìłľ
    -0.08
    POSITIVE LOGITS
    çīº
    0.11
    éĴĹ
    0.09
     oran
    0.09
    åĸľæ¬¢åIJĥ
    0.09
    央
    0.09
    æĶ¾å¼ĥäºĨ
    0.09
    fort
    0.09
    sap
    0.09
    sville
    0.09
    è¦ģæľī
    0.09
    Activations Density 0.006%

    No Known Activations