© 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. 26780
    Prev
    Next
    INDEX
    Explanations

    return values

    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
    çĥŁ
    -0.10
    èµ·æĿ¥äºĨ
    -0.09
    etty
    -0.09
    éĢĨè¡Į
    -0.09
    å¦Ĩ
    -0.09
    æ¡ĥ
    -0.09
    inger
    -0.09
    _GPU
    -0.08
     peg
    -0.08
     bis
    -0.08
    POSITIVE LOGITS
    (Return
    0.11
    (return
    0.10
    纥
    0.10
     uwagÄĻ
    0.09
    åĽŀæĿ¥
    0.09
     kepada
    0.09
    atab
    0.09
    uce
    0.09
    常æĢģ
    0.09
    åī¯æĢ»è£ģ
    0.09
    Activations Density 0.092%

    No Known Activations