© 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. 4207
    Prev
    Next
    INDEX
    Explanations

    invoking, inv, invisible, invitational, invincible

    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
    çĶ¨äºº
    -0.09
    è¾Ī
    -0.09
    对她
    -0.09
    adow
    -0.09
    .abort
    -0.08
    æĹĸ
    -0.08
     retr
    -0.08
    isan
    -0.08
    éĥ½è§īå¾Ĺ
    -0.08
    IAN
    -0.08
    POSITIVE LOGITS
    ersed
    0.11
     lou
    0.09
    ระยะ
    0.09
    大èĥĨ
    0.09
    uts
    0.09
    hou
    0.09
    ibly
    0.08
    egen
    0.08
    æIJª
    0.08
    ible
    0.08
    Activations Density 0.089%

    No Known Activations