© 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. 54292
    Prev
    Next
    INDEX
    Explanations

    permission granted

    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
    ToFit
    -0.10
    å±¥
    -0.10
    æ§Į
    -0.09
    ile
    -0.09
    gment
    -0.09
    phis
    -0.09
     gim
    -0.08
    æī©åħħ
    -0.08
    çIJī
    -0.08
     Gim
    -0.08
    POSITIVE LOGITS
    èłķ
    0.10
    èĮĦ
    0.10
    èĩªæĿ¥
    0.09
    æĺ¯æ¯Ķè¾ĥ
    0.09
    ÑĪиб
    0.09
     вÑĭпÑĥÑģк
    0.08
    omatic
    0.08
    lessly
    0.08
    毯
    0.08
     Brook
    0.08
    Activations Density 0.021%

    No Known Activations