© 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. 2300
    Prev
    Next
    INDEX
    Explanations

    describe how or what

    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
    çĦ¶æĺ¯
    -0.11
    ìį¨
    -0.10
    eral
    -0.09
    æĴŃæĬ¥
    -0.09
    PEAR
    -0.09
    OTH
    -0.09
     окÑĢÑĥг
    -0.09
    yt
    -0.09
    è¢Ħ
    -0.09
    _reload
    -0.08
    POSITIVE LOGITS
    ä¸įåĩºæĿ¥
    0.11
     how
    0.09
    ives
    0.09
    éģĤ
    0.08
    ively
    0.08
    壮
    0.08
     cómo
    0.08
    ble
    0.08
    atch
    0.07
    oriented
    0.07
    Activations Density 0.060%

    No Known Activations