© 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. 37841
    Prev
    Next
    INDEX
    Explanations

    first thoughts and reactions

    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
    伺
    -0.10
    tsy
    -0.10
    okies
    -0.08
    åıĮåIJij
    -0.08
    æħİ
    -0.08
    HEST
    -0.08
    okie
    -0.08
    /slick
    -0.08
    tering
    -0.08
    vre
    -0.08
    POSITIVE LOGITS
    æĥ³åĪ°äºĨ
    0.14
    æĥ³åĪ°
    0.12
    ç«ĭåĪ»
    0.12
    èĦijæµ·ä¸Ń
    0.11
    èѬ
    0.10
    è¿ħéĢŁ
    0.10
     immediately
    0.10
    ç«ĭåį³
    0.10
    èĦijæµ·
    0.10
    ç«ĭ马
    0.10
    Activations Density 0.048%

    No Known Activations