© 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. 55388
    Prev
    Next
    INDEX
    Explanations

    vague, vague ideas, vague terms

    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
    âĦĭ
    -0.09
    æľ¬äºº
    -0.09
    true
    -0.08
    al
    -0.08
    rek
    -0.08
    ts
    -0.08
    ç¿ģ
    -0.08
    ized
    -0.08
    obuf
    -0.08
    ัà¸ģ
    -0.08
    POSITIVE LOGITS
    ãģĮãĤī
    0.11
    åľ°è¯´
    0.11
    -ish
    0.10
    0.10
    ities
    0.10
    ä¸įæ¸ħ
    0.10
     enough
    0.10
    è®°å¾Ĺ
    0.10
    家ä¼Ļ
    0.09
    ifying
    0.09
    Activations Density 0.007%

    No Known Activations