© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36087
    Prev
    Next
    INDEX
    Explanations

    results of

    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
    åĿļæĮģ以
    -0.11
    ãĥĻãĤ¹ãĥĪ
    -0.10
    erno
    -0.10
     обÑĢаÑī
    -0.09
    çŀij
    -0.09
    çĸ¹
    -0.09
    ily
    -0.09
    iliary
    -0.09
    çıı
    -0.09
    .setVertical
    -0.09
    POSITIVE LOGITS
    丰ç¡ķ
    0.13
    åı¯æĥ³èĢĮçŁ¥
    0.12
    quo
    0.11
    ats
    0.09
    /results
    0.09
    ç»ĵæŀľ
    0.09
    uten
    0.09
    mond
    0.08
    èIJ§
    0.08
    oure
    0.08
    Activations Density 0.141%

    No Known Activations