© 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. 62156
    Prev
    Next
    INDEX
    Explanations

    discrepancy between things

    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
    女è£ħ
    -0.09
    ç¿ĺ
    -0.09
    æĮijéĢī
    -0.09
    _VARS
    -0.09
    uards
    -0.08
    对æĪij们
    -0.08
     Scalars
    -0.08
    encode
    -0.08
    ivot
    -0.08
    æ¡ij
    -0.08
    POSITIVE LOGITS
    å®ŀéĻħæĥħåĨµ
    0.11
     ìĭ¤ìłľ
    0.11
     reality
    0.11
     Actual
    0.10
    ismatch
    0.10
    å®ŀéĻħ
    0.10
     mismatch
    0.10
    éĽ¢ãĤĮ
    0.09
    ä¸Ģèĩ´
    0.09
     Reality
    0.09
    Activations Density 0.223%

    No Known Activations