© 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. 55654
    Prev
    Next
    INDEX
    Explanations

    say nothing at all

    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
    åı£ä¸Ń
    -0.08
     Ye
    -0.08
    åı£
    -0.08
    åı£è¯Ń
    -0.07
    éĴ¦
    -0.07
    rets
    -0.07
    å®Ĺ
    -0.07
     REQUEST
    -0.07
     flo
    -0.07
    aga
    -0.07
    POSITIVE LOGITS
    å¹½
    0.10
    å®ŀä¹ł
    0.10
    etri
    0.09
    IER
    0.09
    éĤ£äºº
    0.09
    iert
    0.09
    å«ĸ
    0.09
    åĨįåĬłä¸Ĭ
    0.09
    éī´
    0.09
     ÙħÙĨÙĩ
    0.09
    Activations Density 0.163%

    No Known Activations