© 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. 63470
    Prev
    Next
    INDEX
    Explanations

    chosen or selected for an honor

    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
    æľªæĪIJ
    -0.09
    rh
    -0.09
    æĵģ
    -0.09
    _ASS
    -0.08
    ç·£
    -0.08
    ä¹İ
    -0.08
    oce
    -0.08
    ä¸įå°½
    -0.08
    åı¦æľī
    -0.08
    peat
    -0.08
    POSITIVE LOGITS
    ity
    0.11
    çľ¼çIJĥ
    0.10
    è¯Ħå§Ķ
    0.09
    çĢĽ
    0.08
    ä¸Ģä¸Ŀ
    0.08
    bens
    0.08
    åĪĹ为
    0.08
    OrElse
    0.08
    çļĦä¸Ģåijĺ
    0.08
    人éĢī
    0.08
    Activations Density 0.031%

    No Known Activations