© 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. 20131
    Prev
    Next
    INDEX
    Explanations

    or followed by fragments

    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
    ëģĶ
    -0.11
    ®
    -0.10
    İ
    -0.09
    odore
    -0.09
    nger
    -0.09
    emia
    -0.08
    emos
    -0.08
    ãģ®ãģĮ
    -0.08
    ether
    -0.08
     Oper
    -0.08
    POSITIVE LOGITS
    chestra
    0.17
    phans
    0.16
    ignal
    0.16
    iginal
    0.16
    chest
    0.14
    leans
    0.14
    naments
    0.13
    IENTATION
    0.13
    thon
    0.13
    acles
    0.12
    Activations Density 0.102%

    No Known Activations