© 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. 30557
    Prev
    Next
    INDEX
    Explanations

    words followed by another word

    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
    çļĦçħ§çīĩ
    -0.08
    enso
    -0.08
    åĦ¿
    -0.08
     imper
    -0.08
    çijŁ
    -0.08
    รว
    -0.08
    ennes
    -0.07
    ä¼ĹæīĢ
    -0.07
    DetailsService
    -0.07
    rances
    -0.07
    POSITIVE LOGITS
    çĪ·
    0.13
    åłĤ
    0.12
     Lounge
    0.11
    ware
    0.10
    ilate
    0.10
    .fm
    0.09
    æĿ¿
    0.09
    ç³
    0.09
    conde
    0.09
    mail
    0.09
    Activations Density 0.297%

    No Known Activations