© 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. 62295
    Prev
    Next
    INDEX
    Explanations

    momentary, acceptable, ordinary, reach, all, plans, control, purification, bring

    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
    asting
    -0.09
    è§Ĩ
    -0.08
    ç¶ļ
    -0.08
    糸
    -0.08
    éĢĢç¨İ
    -0.08
    ìľ
    -0.08
    =create
    -0.08
    fare
    -0.08
    å¤©åľ°
    -0.08
    ary
    -0.08
    POSITIVE LOGITS
    _delegate
    0.10
    æĸ°é²ľ
    0.10
    mah
    0.09
    æĥ³çŁ¥éģĵ
    0.09
    çľĭ书
    0.09
    ç³Ļ
    0.09
    æĺ¯æĪijçļĦ
    0.09
    bounce
    0.09
    åºĸ
    0.09
    çĿ¡åīį
    0.09
    Activations Density 0.263%

    No Known Activations