© 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. 54515
    Prev
    Next
    INDEX
    Explanations

    huge

    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
    ´
    -0.10
    sp
    -0.10
    lor
    -0.10
    son
    -0.10
    neo
    -0.09
     usual
    -0.09
    ss
    -0.09
    ateur
    -0.09
     sheer
    -0.09
    ãģĦãģ¾ãģĻ
    -0.09
    POSITIVE LOGITS
    rams
    0.12
    æ·Ģ
    0.12
    itarian
    0.12
    antry
    0.11
    岬
    0.11
    åµĺ
    0.10
    itecture
    0.10
    ç§ģæľį
    0.10
    raphic
    0.10
    yla
    0.10
    Activations Density 0.008%

    No Known Activations