© 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. 49802
    Prev
    Next
    INDEX
    Explanations

    units of measurement

    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
    æĤ²
    -0.08
    itud
    -0.08
     Misc
    -0.08
    oder
    -0.08
     miscellaneous
    -0.08
    аÑĤÑĮÑģÑı
    -0.08
    éĹ¾
    -0.08
    [arg
    -0.08
    ç¼Ŀ
    -0.08
    çĿ¦
    -0.08
    POSITIVE LOGITS
    å°¤
    0.10
    éªĵ
    0.10
    çĶļèĩ³
    0.10
    éģįåıĬ
    0.10
    è§Ħ模
    0.09
    uis
    0.09
    èı½
    0.09
    é¹Ń
    0.09
    åłª
    0.09
    пÑĥб
    0.09
    Activations Density 0.050%

    No Known Activations