© 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. 33635
    Prev
    Next
    INDEX
    Explanations

    outward and inward directions

    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
    ãĥ¼ãĥ«
    -0.10
    çĬ¶åĨµ
    -0.10
    oyer
    -0.10
    TURE
    -0.10
     доÑģÑĤ
    -0.10
    è´»
    -0.09
    æݪ
    -0.09
    zent
    -0.09
    ifer
    -0.09
    å¾ĹäºĨ
    -0.09
    POSITIVE LOGITS
    mente
    0.14
    ly
    0.14
    robe
    0.11
    ness
    0.11
    /out
    0.09
    sWith
    0.09
    ÑģÑĤв
    0.09
    enheim
    0.09
    ment
    0.08
    _codegen
    0.08
    Activations Density 0.030%

    No Known Activations