© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36436
    Prev
    Next
    INDEX
    Explanations

    append or appendixes

    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
    箪
    -0.10
    稿
    -0.09
    ä¸ĢéĿ¢
    -0.09
    å¦ĤæľŁ
    -0.09
    ogo
    -0.09
     вообÑīе
    -0.09
    顽
    -0.09
    UCK
    -0.08
    uck
    -0.08
    åij¨åĽ´
    -0.08
    POSITIVE LOGITS
    è¿IJèIJ¥åķĨ
    0.09
    -append
    0.09
    enda
    0.09
    象
    0.09
    翼
    0.09
    ãĥĮ
    0.08
     primes
    0.08
    _append
    0.08
     Hedge
    0.08
     append
    0.08
    Activations Density 0.021%

    No Known Activations