© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20660
    Prev
    Next
    INDEX
    Explanations

    qualify as or for

    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
    æīĵç®Ĺ
    -0.10
    ationally
    -0.09
    _blocking
    -0.08
    ledo
    -0.08
    esting
    -0.08
    çīĮåŃIJ
    -0.08
    çªł
    -0.08
    sWith
    -0.08
    plementation
    -0.08
    çºĮ
    -0.08
    POSITIVE LOGITS
    è¨Ĺ
    0.10
    wd
    0.09
    为æŃ¢
    0.09
    ç»ĨåĪĻ
    0.09
     Ñģобой
    0.09
    èģĶ
    0.08
    utos
    0.08
    /dis
    0.08
    ä¸įäºĨ
    0.08
     EITHER
    0.08
    Activations Density 0.010%

    No Known Activations