© 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. 4836
    Prev
    Next
    INDEX
    Explanations

    without warranties or conditions

    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
    idel
    -0.09
    ial
    -0.08
     bottom
    -0.08
    赤
    -0.08
     rav
    -0.08
     Holl
    -0.08
    çĶĺ
    -0.08
    gi
    -0.08
    ULL
    -0.08
    çķĻå®Ī
    -0.08
    POSITIVE LOGITS
    rende
    0.10
    è¿ĩçļĦ
    0.09
    Ñī
    0.09
    ä¸Ģç«Ļ
    0.09
    管çIJĨå±Ģ
    0.09
    ofs
    0.09
    uce
    0.09
    èĤ©
    0.09
    iran
    0.09
    _serialize
    0.09
    Activations Density 0.008%

    No Known Activations